Intended for healthcare professionals

Head To Head

Should smoking in outside public spaces be banned? No

BMJ 2008; 337 doi: (Published 12 December 2008) Cite this as: BMJ 2008;337:a2804
  1. Simon Chapman, professor of public health
  1. 1University of Sydney, Sydney, Australia
  1. sc{at}

    After success in stopping smoking in public buildings, campaigns are turning outdoors. George Thomson and colleagues (doi:10.1136/bmj.a2806) argue that a ban will help to stop children becoming smokers but Simon Chapman believes that it infringes personal freedom

    Indoor smoking bans draw their ethical authority from extensive research showing harm from prolonged and repeated exposures in homes and workplaces, over many years. By contrast, recent agitation to extend bans to outdoor settings like parks, car parks, beaches, and streets is supported by flimsy evidence. Brief exposures to others’ smoke can produce measurable physiological changes.1 2 However, acute exposure studies typically define brief as 15 to 30 minutes—considerably more than usual smoky encounters outdoors.3

    A recent paper concluded that outdoor smoke is rapidly attenuated but for those within half a metre of multiple smokers “between 8 and 20 cigarettes smoked sequentially could cause an incremental 24-hour particle exposure greater than . . . the 24-hour EPA [US Environmental Protection Agency] health-based standard for fine particles.”4 The authors referred to bar patios as where this might happen but state that “sitting next to a smoker on a park bench” might produce such exposure, despite also stating that multiple smokers are required to get to levels that challenge the EPA standard. “Multiple smokers” are rarely seated on park benches next to non-smokers for the time it would take to smoke 8-20 cigarettes.


    Some are affronted by the mere sight of smoking. Others have an evangelical mission to use paternalistic “tough love” to help others quit. Prohibitions on personal behaviours can be justified by the right to interfere with the liberty of people to harm to others. But paternalism is most odious when used as a justification for limiting the choices that adults make when they put only themselves at risk.5 Health facilities banning smoking outdoors often justify this as normative role modelling. This is ethically unproblematic for staff who are contractually obligated to observe employers’ policies but represents ethically muddled thinking when it comes to patients and visitors, who are not somehow “owned and controlled” by health authorities. If they harm no one else by smoking outdoors, they ought not be coerced into signing up to the health promotion values of a hospital when visiting.

    Many smokers support paternalistic policies designed to discourage their smoking. But we do not evaluate the ethics of public health by the willingness of people to give up their autonomy, nor with the success of commandments to obey laws. The ethics here is about respect for the autonomy of individuals to act freely, providing their actions do not harm others.

    There are few differences between the chemistry of tobacco smoke and that generated by incomplete combustion of any biomass: leaves, campfires, petrol, or barbecued meat.6 7 Secondhand smoke is not so uniquely noxious that it justifies extraordinary controls of such stringency that zero tolerance outdoors is the only acceptable policy. Park barbecues aren’t banned for the obvious reason that the amount of smoke involved is trivial. Zero tolerance of tobacco smoke in outdoor public settings is nakedly paternalistic.

    Problems of health argument

    Advocates for smoke-free outdoor areas include those who passionately attest to being severely affected by even the tiniest exposure to smoke. If public health policy is to be evidence based, such claims need to be subjected to scientific assessment.

    Two reviews examined evidence for both the toxiogenic hypothesis (that intolerance of low levels of any environmental agent explains symptoms either through toxicodynamic pathways or by sensitising neural pathways) and the psychogenic hypothesis (that idiopathic environmental intolerance is a culturally learnt phenomenon characterised by an overvalued idea of toxic harm explained by psychological or psychosocial processes).8 9 The reviews concluded that none of the Bradford-Hill criteria for causation10 were satisfied by the toxiogenic theory, but that all of the criteria were met for the psychogenic theory.

    Governments often regulate citizens’ conduct to reduce nuisance, regardless of whether it affects health. Public health research is debased when it lends bogus credibility to what are essentially matters of community preference. If authorities wish to stop smoking on beaches to reduce litter, they should frame their actions in terms of litter reduction, not public health. Landlords wanting to prevent smokers from renting apartments because of complaints about smoke drift from other residents, should be at liberty to do so, but need not invoke public health justifications.

    In most of the world smoking remains a normal, unremarkable, and unregulated activity. Health workers in those nations are desperate to convince governments of how reasonable it should be to remove involuntary tobacco smoke exposure in occupational and indoor public settings. They marshal evidence about disease caused by long term exposure and staunchly defend the credibility of that evidence from the predations of the tobacco and hospitality industries, intent on exposing those risks as trivial. Opponents of clean indoor air will point to campaigns against outdoor smoking and argue that advocates actually want to ban smoking everywhere. Such views are likely to undermine the credibility of advocacy for evidence based smoke-free policies to the great detriment of hundreds of millions of citizens.


    Cite this as: BMJ 2008;337:a2804


    • This article is adapted from Chapman S. Going too far? Exploring the limits of smoking regulations. William Mitchell Law Review 2008;34:1605-20.

    • Competing interests: None declared.


    View Abstract