Smoking in cars is highly dangerous to the smokers and others. The resulting fine particulate matter levels can be as high as in a smoky pub,1 2 and much higher than World Health Organization (WHO) recommended guidelines.3 Laws requiring smoke-free cars when children are present have been recently adopted in a number of North American and Australian jurisdictions. These include South Australia (2007; fig 1),4 Tasmania (children under 18, 2008),5 New South Wales (under 16, 2008)6 Nova Scotia (under 19),7 Yukon,8 and Ontario (in the latter the law came into force in January 2009).9

The Queensland government is in the process of passing a law.10 Five US state or territorial jurisdictions have banned smoking in cars if children are aged under 13, or where there are children in child safety devices: Arkansas, Louisiana, Puerto Rico (all 2006),11^–13 California (2007)14 and Maine (which has a warning from September 2008, civil violation from September 2009).11 12 14^–17 In addition, Kuwait has a law to explicitly prohibiting smoking in any vehicle while driving.18 A wider number of jurisdictions have also banned smoking in some or all public or work vehicles, for instance in Chile,19 Germany,20 and England.21

Little evaluation of such laws has been reported. Keith Evans, Chief Executive of the South Australia Drug and Alcohol Service was reported as saying that the police there have cooperated “particularly well” with this public health move. In the first year of the law 125 drivers were fined, and 38 cautioned, although the overall aim of the law was not to fine significant numbers, but to “gradually bring about a change in behaviour”.22

In other jurisdictions the focus has been on mass media campaigns to promote voluntary smoke-free cars. This has included government and non-government media campaigns in Canada,23 New Zealand,24 New South Wales,25 and Western Australia.26

Knowledge of the harm from second-hand smoke (SHS) exposure in cars may be a driver of attitudes. There is limited research on public knowledge about the SHS hazard from smoking in cars. In the USA, 77% of a national survey, when asked how much smoking in a car affects the health of children, answered either “a lot” or “a great extent”.27 Residents of rural counties in the USA were slightly less likely to recognise that smoking in a car is harmful to children (approximately 75%) than residents of urban counties (approximately 80%).28

The need for smoke-free car laws is indicated by the level of reported exposure to SHS in cars (by smokers and non-smokers). Surveys of adults, in Canada and New Zealand during circa 2003–2007 have shown very similar prevalence of exposure to smoking in cars: between 22% and 26%.29^–34 Surveys since 2000 of students, in Wisconsin (2000–2003),35 Nebraska (2002 and 2006),36 New Zealand (2006),37 and Canada (2004 and 2005–2006)38 39 have shown a wider range of prevalence of exposure to smoking in cars: between 24% and 48%. However, except for the exposures found in Wisconsin (43%) and in Nebraska (48% and 40%), all the exposure prevalences were between 28% and 24%. In a 1996 survey in The Netherlands on the exposure of infants to SHS in cars, 8% of parents reported that their infant had been exposed to tobacco smoke in a car in the previous 7 days.40

The context of attitudes to smoke-free car laws includes the tensions between health promotion, health protection and concerns about excessive state paternalism.41 42 The commercial opponents of such interventions can exploit such concerns.43 44

We expect that smoke-free cars is likely to be an area for ongoing health policy development, due to the very high levels of SHS that can occur in the car environment,45 46 and the resulting political response to visible, avoidable harm to children.47 However, there appears to have been no published reviews to date of public attitudes to smoke-free car laws.

This brief review considers surveyed attitudes to legal restrictions around smoking in cars.

METHODS

We undertook Medline and Google Scholar searches for the period prior to mid-November 2008 for English language sources, and followed-up subsequent references. A range of search terms were used, but particularly: “smoke, smoking or tobacco”, “car or vehicle”, “ban”, “attitudes” and “law, legislation or regulation”. Data on national and subnational jurisdictions (eg, states and provinces) were collected, but not for smaller administrative units, eg, cities or councils.

RESULTS

A total of 15 relevant studies were identified (table 1) with data from 10 jurisdictions (Canada, UK and New Zealand, and from 1 Canadian province, 4 Australian states and 2 US states). The major division found in the type of attitude found stemmed from whether children were considered in the survey question.

One study was only of smokers’ attitudes (in New Zealand 2007–2008),48 and one study (from a large Canadian sample) was of youth attitudes.38 Only one study (from South Australia) reported attitudes after the implementation of a smoke-free car law.49

The surveys used by four studies, using surveys from 1988 to 2004 in Indiana and New Zealand, did not mention children. The surveys in the other 11 studies, in 1994, 1995 and from 2000 to 2008 (in New South Wales, Ontario, Canada, Victoria, California, New Zealand, South Australia, Queensland and the UK) specifically asked about attitudes if children were present.

This difference in the question is reflected by a sharp increase in the approval of smoke-free car laws in those jurisdictions, compared to previous surveys in the same or similar jurisdictions. Where children were not mentioned, the results favouring no smoking in cars during 1988–2003, in those areas of Australasia and North America where there were surveys, ranged from 28% (for no smoking by car passengers in Indiana in 2003) to 47% (New Zealand, 1988). A 2004 New Zealand survey found only 40% in favour of no smoking in cars, but 76% disagreed that it is “okay” to smoke around non-smokers inside cars even when there are windows down.

The 1994 and 1995 surveys in New South Wales that mentioned children in the question found 72% and 55% support for a smoke-free car law respectively.50 51 Then, in five of the eight surveys in Canada (2004),38 Victoria (2004–2006),52 New South Wales (2000, 2004),53 54 California (2005),55 New Zealand (2007–2008),48 South Australia (2007, 2008),49 and UK (2008),56 there were support levels from 81% to 96% for no smoking in cars with children. The exceptions were 57% and 55% support in New South Wales in 2000 and 2004,53 54 and 76% support in the UK in 2008.56

Smokers’ attitudes

There were 11 studies that reported the attitude of smokers to smoke-free car laws. While they generally were less likely to support smoke-free car laws, one survey in New Zealand in 2007–2008, which asked about cars containing preschool children, showed the highest level of support in this review, at 96%.48

Smokers’ attitudes also varied depending on whether children were included in the question. In the three studies when children were not included (reporting on two surveys in New Zealand in 1988 and 2004,57 58 and one in Indiana in 2003),59 support for smoke-free cars varied between 6% and 23%. In the eight studies where children were included in the question, support for smoke-free cars varied between 45% (New South Wales in 2000)53 and 96%.48 The latter specified “preschool children” in the question. The five studies from surveys in 2005 or since (in Victoria, California, Queensland, New Zealand and South Australia) reported smoker support of 77% or over for smoke-free cars containing children.48 49 52 55 60

Increases in support over time

There was some indication of an increase over time in supportive public attitudes to smoke-free car policies. Examples of this include survey data (using the same question) showing an increase in New Zealand support for laws on smoke-free cars from 29% in 1999 to 41% in 2003,61 and Victorian support from 90% to 92% during 2004–2006. An exception to the trend may have occurred in New South Wales (see below). The support for laws for smoke-free car containing children, in surveys of adults up to 2004, ranged from 55% to 72%. In contrast, in surveys since 2005 support has ranged from 76% to 96%.

Attitudes by ethnicity, education and income

The only surveys found with data on attitudes to smoke-free car laws by ethnicity were from New Zealand and California. In California in 2005, support from Hispanic (97%) Asian (95%) and African Americans (94%) was stronger than for non-Hispanic whites (88%).55 In New Zealand, Maori (indigenous) support for no smoking in cars was consistently much higher than that of the whole population during 1999–2003.61 The 2005 California survey also showed a gradient of support by education and income, with the least educated (96%) and those with the lowest income (98%) giving greater support than the most educated (90%) and those with the highest income (91%).55

DISCUSSION

The most notable feature of the results is the effect that the consideration of children or others has on the level of support for smoke-free car laws. When respondents are thinking of children, support for laws banning smoking in cars has been high (76% or over) in the six jurisdictions surveyed since 2005. There is some indication that public attitudes to government smoke-free car laws have grown more favourable over time, and this may have been as smoke-free policies for indoor environments have increased in scope.

Another feature of the results is the evidence that in much of North America, the UK and Australasia, majority public opinion is unlikely to be a barrier to passing relevant laws. This indicates that concerns about the “invasion” by government into “private” space do not necessarily apply in this case, particularly where children are involved.47

Limitations

When considering the generalisability of the findings in this review to developed countries in general, a major limitation is that data come from jurisdictions in only 6 different countries (a total of only 10 different jurisdictions, all predominantly English speaking). Furthermore, for general population attitudes, there are data from only two nations (New Zealand and the UK) although there is youth data for all Canada. Within Canada the general population data are for only Ontario (38% of the national population) and within the US only for Indiana and California (14% of the national population). The Californian data is unlikely to be representative for the USA, as attitudes to smoke-free laws in California are probably more progressive than those of the USA in general, given that this state is a leader in tobacco control. There is also a wide range in the survey periods and in the survey methods used, across the 16 studies, which limits the ability to generalise even across the limited number of jurisdictions.

Variations related to survey methods

There appear to be wide variations of support for smoke-free car laws even within the same jurisdiction, due to variations in survey question, sample type, or other methodological variation. For instance, results from surveys in New South Wales varied from 72% support for a law for smoke-free car containing children in 1994,50 to 57% in 2000 among car owners,53 and 55% in 2004.54 The apparent decline of support was present among non-smokers, as well as in the whole survey samples.

Research implications

Given the hazard of SHS in the car setting, and the relatively high levels of public support for smoke-free car laws, this would appear to be an area in which further research (to inform policy development) is desirable. The priority is for surveys in jurisdictions for which there exists no data on attitudes to smoke-free car laws. Care with question design is important, if comparability with leading tobacco control jurisdictions is desired.

Another research gap is the lack of evaluation data on post-law smoke-free car interventions around public acceptability and compliance. Such research is now possible, since a number of jurisdictions have recently adopted such laws (and states such as California have a strong record for monitoring tobacco control activities).55

There is also no survey series (with the same question) that provides smoke-free car law attitude data from before and after any smoke-free legislation (eg, for other public places). Such a series may help gauge the collateral impacts of such laws. Repeat surveys will also enable a focus on the rate and direction of change, rather than just the level of support.

Policy implications

Given the experience with other smoke-free laws, evidence of public and smoker support for smoke-free car laws would make the introduction of them much easier for policymakers in democracies.62 63 Attitudinal shifts among the public and smokers towards smoke-free car law adoption, and to compliance with these laws, may be more likely to occur if: (i) awareness of the SHS health hazard in their country improves; (ii) attitudes towards protecting children from harm strengthens; and/or (iii) their country enhances tobacco control activities in general. In particular, smoke-free cars may be more likely if legal restrictions are introduced on smoking in a range of other settings, such as workplaces and hospitality settings. For instance, smoke-free school policies may alert parents to the need to protect their children from SHS in cars (and in homes).

However, because of the likely high public support for the particular intervention of smoke-free car laws in many developed countries, it may be able to be adopted ahead of, or at the same time as, other government smoke-free policies. The examples of Arkansas, Louisiana and Puerto Rico support this possibility.

Cars are the settings in which children are most likely to be exposed to the highest levels of SHS, and where the children have no easy way of avoiding it. There are therefore strong ethical and public health arguments for government interventions, such a mass media campaigns for voluntary policies, and for passing smoke-free car laws.43 45 46 64 65 Compared to other smoke-free legislation (for instance, for smoke-free bars) there is likely to be less opposition to laws for smoke-free cars containing children, as few policymakers wish to be seen as not wanting to protect children.47

Even if smoke-free car laws appear to some to politically difficult in some jurisdictions, there may be obligations on governments to act to protect children.66 We suggest that in most, if not all jurisdictions, smoke-free car laws are required to protect significant numbers of children. Social marketing (eg, media campaigns) will only be effective in protecting some. Those most socioeconomically deprived may be the least likely to be protected by social marketing campaigns.67^–69 Adults need protection from SHS as well, and governments which aspire to protect vulnerable sections of their populations need to also consider smoke-free car laws which protect people of all ages.