Cigarette smoke is a toxic mixture containing around 4000 different chemicals including a range of carcinogens, irritants, and toxins.¹ It is therefore no surprise that inhaling cigarette smoke, either actively as a cigarette smoker or passively through exposure to exhaled and sidestream smoke from other smokers, is bad for health.¹ In fact, active smoking kills more people in economically developed countries than any other preventable cause, and currently accounts for over 100 000 deaths (or about 20% of all deaths) each year in the UK.² Worldwide the annual death total is currently close to 5 million.³ Passive smoking is also a major problem, causing at least 12 000 deaths each year in the UK,⁴ or about 2% of the current annual total, and probably a similar proportion in other developed countries. On these figures alone it should be clear that we have sufficient evidence to conclude that preventing exposure to cigarette smoke, either active or passive, should be a fundamental priority of all clinical and public health practice. We already know what the main health effects of smoking are, or in relation to passive smoking, are likely to be. So do we need more research defining the adverse effects of passive smoking in still greater detail?

The answer to this question is that we do, and primarily to inform, support, or otherwise enhance the political process towards effective tobacco control. While governments continue to prevaricate over the implementation of simple public health measures to reduce smoking incidence and encourage smoking cessation, new evidence can help tobacco control lobbyists and supporters both inside and outside the government and health services to press for change. New evidence is helpful in keeping the media interest in a topic alive and thus maintaining pressure on policy makers to act. In parallel to the political process, much that is good for public health can come from decisions made in courts of law. If the emergence of new evidence on tobacco effects were to contribute to successful legal claims for injury arising from tobacco smoke exposure, for example at work, that would also probably have a major impact on the measures employers deem necessary to take to protect employees from passive smoke exposure. New evidence therefore matters but, to be useful, it should ideally be strong enough to contribute to the above processes.

In this issue of Thorax Eisner and colleagues report new data suggesting that passive smoking exacerbates asthma.⁵ There is already an extensive literature, reviewed elsewhere, showing that passive smoking is associated with an increased risk of asthma in children and adults,^1,6–8 and observational evidence (including other work by Eisner et al) that passive smoke exposure exacerbates symptoms in people with asthma.^9–14 These latter findings are supported by the results of experimental challenge studies indicating that passive smoke exposure has adverse effects on airflow and/or airway responsiveness in asthma.^15–19 Collectively, therefore, the existing evidence provides strong grounds to suspect that passive smoke exposure is bad for asthma. What this new study argues is that the existing epidemiological data are based predominantly on self-reported passive smoke exposure, and as such may be biased, and instead uses personal nicotine monitoring and hair nicotine and cotinine assays to provide objective measures of passive smoke exposure in a cohort of patients recently discharged from hospital with acute asthma.

The study concludes that people in the highest tertile of personal exposure to nicotine have more severe asthma, while those with higher levels of hair nicotine in the past month had a higher risk of hospitalisation for asthma between the index (recruitment) admission and the time of hair sampling (referred to as “baseline risk”). There were other potentially significant findings in the several hypothesis tests carried out across three levels of exposure for five exposure measures and five outcomes presented in tables 5 and 6 of the paper, and the overall pattern of the findings on the relation to symptom scores is consistent with the interpretation that individuals with higher levels of passive smoke exposure had more troublesome asthma. The findings on the risk of hospitalisation were perhaps less convincing. Participation in the study was low, since only 53% of those eligible for inclusion completed an initial telephone interview and less than 25% of those who completed the interview (that is, under 12.5% of the original proposed study population) participated in the exposure measures. The representativeness of this participant population in relation to any association between passive smoking and asthma exacerbation is therefore questionable. By virtue of its observational design, the study is also unable to determine whether the relation between passive smoke exposure and asthma severity is causal or is confounded by other factors. For example, asthma severity is strongly dependent on adherence to appropriate treatment, but access to and use of treatment may be lower in relatively deprived individuals who are also more likely to be smokers or to mix with other smokers both socially and at work.²⁰ The authors have used level of education to control for socioeconomic status but this may be insufficient to deal with this problem. The individual personal smoking status of participants does not appear to have been validated; only current non-smokers were included in the study, but smoking status appears to be based on self-report. Objective measures of serum, salivary, or urinary cotinine levels and/or exhaled carbon monoxide would have been helpful to validate these reports. These and other criticisms mean that, while the study concludes what many with an interest in this area would like or believe to be true, the main findings could still be due to chance or bias.

On the other hand, previous studies using objective markers of passive smoking in children have produced strong evidence that higher levels of exposure are associated with more severe asthma,^21–23 and the experimental challenge data cited above^15–19 provide support for the hypothesis that this is also likely to be true in adults. It is therefore reasonable to conclude that passive smoke exposure is likely to be bad for all people with asthma who would be well advised to avoid exposure, as indeed is the case for everyone else. Whether this latest addition to the evidence base represents an important step in terms of advancing the public, political, and legal process towards effective public protection from the effects of passive and active smoking remains to be seen.