Protecting children from second-hand smoke

Since the 1970s, we have known that second-hand smoke (SHS) makes children sick, but we have only recently begun to quantify tobacco-related ill health, and to monitor the effects of public health interventions. In this issue of the European Respiratory Journal, Been et al. [1] quantify, for the first time, the effects of smoke-free legislation on hospitalisations for paediatric respiratory infections.

In 1974, two Lancet papers found that infants with smoking parents had higher hospital admission rates and higher risks of pneumonia and bronchitis [2, 3]. Since then, an accumulating body of evidence has confirmed a higher incidence of cough, wheeze, asthma, upper and lower respiratory tract infections (RTIs), preterm birth, and sudden infant death [4–6]. Globally, ∼40% of children are exposed to SHS [7]. Oberg et al. [7] conservatively attributed 603 000 deaths (1% of all deaths) and 10.9 million disability-adjusted life years (DALYs) (0.7% of all DALYs) to SHS, based on area-specific prevalence of those exposed, and on disease-specific risk estimates for lower respiratory infections, otitis media, asthma, lung cancer and ischaemic heart disease. For this assessment, the authors considered only the listed outcomes with level “A” evidence; real numbers will be higher. Children accounted for 166 000 (28%) deaths and for 6.6 million (40%) DALYs caused by SHS; and within childhood disease, lower respiratory infections were most relevant. SHS is a group 1 carcinogen with no safe levels, and children with their immature immune system and developing lungs are particularly vulnerable [8].

Parents, who smoke at home or in the car, are responsible for children's exposure to SHS [9]. To improve child health, we therefore need interventions targeted at adults: preventing them from taking up smoking, or helping them quit. One approach is that of individual prevention, which attempts to change parents' attitudes and educate them through individual counselling, education or smoking cessation programmes. A second approach is structural, which depends on changing the environment and organisational structures by methods such as economic incentives, reducing the availability of cigarettes, tobacco-free advertising, or smoke-free public spaces [10].

Individual prevention remains important in clinical practice. Combining medication and counselling by a physician doubles the chance that individual smokers will quit, and is more cost-effective than other clinical interventions [10]. However, absolute rates of quitting remain low and smoking cessation programmes, including nicotine replacement therapy, electronic cigarettes and nicotine vaccines have small effects [11–13]. Even when patients are severely ill with COPD [14], or are offered financial rewards for successful quitting [15], quit rates hardly exceed 15%. Families of lower socioeconomic status (SES), who have a higher prevalence of smoking [16], tend to profit less from individual prevention [17].

When a large proportion of the population is at risk and the target behaviour is common, structural prevention may be more effective, and there exist very effective structural interventions to reduce smoking. Price increases are by far the most effective measure [10, 18–20]. Campaigns designed to tarnish the image of smokers are also effective, and so are regulations that promote smoke-free public places [10]. Despite their effectiveness, these interventions are rarely implemented. To change this, the World Health Organization (WHO) developed the Framework Convention on Tobacco Control (FCTC), the first evidence-based treaty negotiated under its auspices [21]. The FCTC represents a paradigm shift in regulatory strategies to address addictive substances. In contrast with previous drug control treaties, the FCTC emphasises the importance not only of supply issues, but also of demand reduction strategies. The tobacco companies fiercely opposed the FCTC and employed a subtle strategy to defeat smoke-free policies and legislation [22–24], but the convention was adopted by the 56th World Health Assembly on May 21, 2003 and entered into force on February 27, 2005. The 180 countries party to the WHO FCTC recognise that there is no safe level of exposure to tobacco smoke, and recommend effective protection measures. Guidelines for implementing Article 8 specify that smoking and tobacco smoke should be eliminated from indoor work places, indoor public places and public transport. Yet, only 16% of the world's population is currently covered by comprehensive smoke-free laws [21, 25, 26].

For adults, the FCTC's passage in 2005 led to dramatic positive health effects. A meta-analysis from 2012, including 45 studies, showed that comprehensive smoke-free legislation was followed by lower rates of hospital admissions (or deaths) for four diagnostic groups: coronary events (relative risk 0.848); other heart disease (relative risk 0.610); cerebrovascular accidents (relative risk 0.840); and respiratory disease (relative risk 0.760) [27]. More comprehensive laws were associated with greater reduction of risk. Further evidence on effectiveness, particularly on respiratory admissions, was added by subsequent publications [28, 29].

Less research is available on children, who profit indirectly from the FCTC [30–33]. A recent review by Been et al. [34] showed that hospital attendances for asthma (−10.4%) and preterm births (−10.1%) decreased after the introduction of smoke-free legislation. However, data on childhood respiratory infections, which account for most SHS-attributable ill health, were lacking.

In this issue, Been et al. [1] present the first study on smoke-free law and hospitalisation for RTIs in children. They analysed hospital episode statistics to investigate the association between smoke-free legislation introduced in England on July 1, 2007, and hospitalisation trends for upper and lower RTIs from 2001 to 2012 in children <15 years of age. They carried out a careful analysis adjusted for confounders, seasonal variation, temporal autocorrelation, population size changes, and underlying incidence trends. A sensitivity analysis explored potential interference from the introduction of pneumococcal conjugate vaccine. They found that introduction of smoke-free legislation was followed by an immediate drop in RTI admissions (−3.5%), mainly attributable to lower RTIs (−13.8%). Reductions in admission for upper RTIs were lower and incremental. The study estimated that 11 000 fewer children were admitted to hospital every year, resulting in savings of roughly £17 million per year. The gradual decrease (but not the acute effect) varied by SES and was largest among children of lower SES.

The study demonstrates the far-reaching effects of a simple and cheap structural intervention: a smoking ban in workplaces and public places. Children are infrequently in workplaces and public places where smoking is common. The effect of the ban must, therefore, have been mediated via a reduction of SHS in their homes or cars, as parents might have stopped smoking and others refrained from smoking indoors. Several studies have shown that children's SHS exposure dropped after introduction of smoke-free legislation [32, 33]. For example, in the health survey for England, the proportion of smoke-free homes increased from 30% before the smoking ban to 48% afterwards. Likewise, the proportion of children with urine cotinine levels below the detection limit increased from 34% to 41% [32]. These effects are larger than what is usually attained by individual-based stop smoking programmes, and legislation is far cheaper. The effects could have been larger if smoking bans had included cars [35] and had been combined with higher tobacco prices [18–20].

More effective implementation of the WHO FCTC should be a central goal of political and health authorities in Europe. Political engagement of doctors and medical societies, supporting structural policies, will contribute more to reduce tobacco-related health problems in their country than clinical counselling alone (which of course keeps its value). The claim by Schroeder and Warner [36] that “at a time when all eyes are focused on health care reform, escalating medical costs, and childhood obesity, cigarette smoking remains by far the most common cause of preventable death and disability,” is still true and also applies to Europe. Been et al. [1] give us even more evidence that child health can be improved easily, effectively and cheaply, if we want.