Publication
Passive smoking and associated causes of death in adults in Scotland
| Contents: | Study Aim Executive Summary Study Aim Methods Analysis Methods Analysis Discussion Key references Discussion Key references Appendix Executive Summary Appendix |
Discussion
Discussion
Estimation of the number of deaths among adults in Scotland which are caused by exposure to second-hand smoke is not an exact science. Knowledge of how many people in the population exposed to ETS, both currently and historically, is sparse. The risk estimates associated with even broad levels of exposure have had to be based on international studies. Most of these relate to risks among lifelong non-smokers, as the literature concerning ex-smokers is virtually non-existent.
The level of risk experienced in Scotland is a combination of the proportion of the population exposed and the magnitude of the exposure. In particular, the magnitude of exposure refers to that built up over a lifetime. In that respect, we are lucky in Scotland to have studies with some measure of the proportion exposed historically, as this is probably the closest indicator of lifetime exposure we are likely to get. As knowledge about the risks of passive smoking became more widely disseminated in the 1980s and 1990s, there has undoubtedly been a change in the way in which some smokers smoke when in the presence of non-smokers. This will make it difficult to assess more recent ETS exposure. However, current consequences in terms of the major chronic diseases will be much more a function of what happened some 20 to 30 years ago in terms of ETS exposure.
Our estimate of around 850 to 950 deaths per year (among lifelong non-smokers) relates to deaths that have occurred since 2000. It relates to the four causes of death most strongly linked to smoking. Other causes of death have been linked to active smoking and it is plausible that there will be some small additional contribution from these other causes. Passive smoking represents the greatest risk to public health when compared to other forms of 'involuntary' environmental exposure.
A number of assumptions have had to be made. These include both the 'active' and 'passive' distributions of smoking historically in Scotland, the magnitude of the relative risks for each of the diseases considered and the latencies for the different diseases. While the assumptions used have not been ideal, they are genuine observations and it is unlikely that better information relating to the appropriate time period will become available. Sensitivity analysis has been undertaken (see appendix) to investigate how reliant the final estimates are on the assumptions made. It is reassuring to observe that the estimates produced under different assumptions lie within a range of about 10% (from 850 to 950 deaths per year). We have chosen to use the more conservative estimate.
This paper has concentrated on the impact of ETS exposure among lifelong non-smokers. With a substantial number of smokers quitting over the years since the first reports on the harm of active smoking in the 1950s/1960s, there will be a significant risk from passive smoking amongst the ex-smoking group. However, the information required to provide a realistic estimate of their contribution to passive smoking deaths, is virtually nil. We need to know the pattern of ETS exposure among ex-smokers - it is clear from the data from the Renfrew/Paisley cohabitees study that smokers who live with non-smokers are more likely to quit smoking and hence their domestic exposure would be minimal. Also it is not known to what extent smokers who live together might also quit smoking together. There is also the issue of how long it takes for the risk of the main smoking-related diseases takes to 'wash out' , i.e. their risk returns to approximately that of the lifelong non-smoker. Until the risk associated with active smoking can be considered to have returned to a non-smoking level, it is virtually impossible to prescribe a risk associated with passive smoking. For lung cancer, this 'wash-out' time is of the order of 25 years whilst for the vascular causes it is considerably shorter (about five years). There are no studies that have looked at the risk of ex-smokers for the main causes of death due to ETS.
The current trends show that ex-smoking rates among men are considerably higher than they were in the 1970s and that the rates are moderately higher among women. This implies that there are a substantial number of ex-smokers at risk from ETS exposure. While difficult to quantify, there would appear to be as many ex-smoking men and women exposed to ETS as there are lifelong non-smokers and may well be more. On this basis, given that the main contribution to deaths related to ETS exposure comes from the vascular causes and they are the causes which have short 'wash-out' times, then we could be looking at a similar number of ETS-related deaths among ex-smokers as we estimate for lifelong non-smokers.
What impact would a ban on smoking in the workplace have on the number of deaths caused by passive smoking? This would depend on how successful the ban was and how much of an impact it would have on the total ETS exposure for an individual. As a rough guide, a 50% reduction in the average ETS exposure could be expected to deliver a 50% reduction in the number of attributable deaths over time.
One area on which this report does not comment is the impact the introduction of a ban on smoking in the workplace might have on active smoking rates. Well over 10,000 deaths per year in Scotland can be attributed to active smoking and even small reductions in the active smoking rates will be hugely beneficial. The consequences of such a ban should be assessed not only in terms of the number of ETS-related deaths prevented but also its likely impact on active smoking rates in Scotland.