The Invisible Killer and the Problem of Phantom Risk: Some Observations on the UK’s Flawed Risk Management Process

Author: John Luik
Article Published: 24/08/2007

---

A Talk delivered to the Wilton Park Conference Managing Risk: Sensible Precaution or Fear of Trying 27 March, 2007

The advertisement is graphic and memorable, though not, unfortunately, for its truthfulness. It shows guests at a wedding reception enveloped in a dark smoking cloud from a lone smoker’s cigarette. Though the smoker frantically tries to wave the smoke away, the voice over makes it clear that this does not make any difference to this “invisible killer” secondhand smoke.

The advertisement is part of a DOH new campaign which started at the beginning of this month to highlight the dangers of passive smoking for nonsmokers in advance of the July public smoking ban in England. According to Public Health Minister Caroline Flint, speaking about the new advertising campaign, “Smoking is harmful not just to smokers but to the people around them. What this new campaign brings home very clearly is the full impact of second-hand smoke.” What the minister did not mention, but which is the subject of this discussion is how profoundly misleading the government’s advertising campaign is and how indifferent it is to the Lord’s report on risk management.

We wish to examine the public smoking ban discussion in the Lord’s report as a case study on the government’s management of risk. We will argue that:

the debate about public smoking and public smoking ban provides evidence about

1) how risk can be manufactured and manipulated for other ends;

2) how unscientific, indeed, how indifferent to evidence, risk management can be;

3) how insusceptible to remedy the risk process and risk conclusions can be.

I shall begin by recounting what the Lord’s report says about the government’s risk policy on public smoking, next describe what I think is reasonable to conclude about the government’s risk management of public smoking and finally offer some suggestions about how risk management might be improved based on the lessons of managing public smoking.

1)      What the Lord’s Found

According to the Lord’s the evidence suggests that the Government has a “sound and potentially effective framework” for risk assessment. The crucial question is whether this framework is properly applied in the management of risk. Three principles outlined in The Green Book produced by HM Treasury, are central to legitimate risk policy:

1)      Openness and transparency

2)      Proportionality and consistency

3)      Evidence-based

Tony Bandle, from the Health and Safety Executive, yesterday argued that these principles are strictly followed in risk management. As we shall see, the evidence from the Lords report, particularly with respect to the HSE, suggests otherwise. Let’s begin with the question of whether the Lords found the government’s public smoking risk management policy was evidence-based.

The most crucial initial questions in terms of risk management are is there a risk and how significant is that risk? One’s management strategy obviously is contingent on the significance of a risk which in turn depends on its measurability.

The Lord’s heard evidence from a variety of sources about whether public smoking was a risk and if so, how significant a risk it was. Of particular note was OxfordUniversity epidemiologist and campaigner against ETS Sir Richard Peto who suggested that the risks from public smoking are “uncertain and unlikely to be large.” (Section 77) As Gio Gori will argue latter the crucial component of science- measurability- is according to Sir Richard, absent from public smoking risks. Indeed, contra the government’s claims that public smoking kills, Sir Richard said that he did “not want to be case in the role of advocating banning smoking in public places or in private places.”

The reason for Sir Richard’s hesitancy to argue that public smoking is a substantial risk to the health of nonsmokers is to be found in the epidemiological evidence about secondhand smoke. This evidence suffers from a number of crucial problems, most importantly that it is based on the decades old recall of nonsmokers about the frequency and extent of their exposure to secondhand smoke. Nevertheless, of the 75 published studies on secondhand smoke and lung cancer involving nonsmoking spouses, only 14 show a statistically significant correlation between ETS exposure and lung cancer. Of the 42 studies on secondhand smoke and heart disease in nonsmokers, 18 show a statistically significant correlation between ETS exposure and heart disease. Again, in terms of workplace exposure to public smoking, of 24 studies 19 are not statistically significant. Moreover, in the studies which are statistically significant, none of the relative risks exceed 2 which is considered a weak risk. As the National Cancer Institute noted in 1994 “In epidemiologic research, [increases in risk of less than 100 percent] are considered small are are usually difficult to interpret. Such increases may be due to chance, statistical bias, or the effects of confounding factors that are sometimes not evident.”

Clearly then, the Lord’s found that the epidemiological evidence did not support the conclusion that passive smoke was a significant health risk to nonsmokers, contrary to the Government’s claim that passive smoking imposes a significant health risk on workers and others exposed to environmental tobacco smoke” (Section 76), and contrary to the DOH advertisement which describes ETS as an invisible killer.