Feb 17 2017

The Science of Smoking Bans

smoking-banIn a recent article for Slate, Jacob Grier argues that the science used to justify widespread bans on smoking in public places was flawed. Recent more robust research has show little to no health benefit from such laws, he argues. While he has a point regarding the arc of scientific evidence, I think he is going too far in the other direction in his conclusions about the science.

Second Hand Smoke

The current consensus of evidence is that there are health risks to second hand smoke, although they are statistically small. Debate centers around the magnitude of the effect, with few doubting that there is a negative health effect. Negative health effects include heart attacks, lung cancer, stroke, and exacerbation of asthma. On a population level, even small increased risks result in large numbers of excess deaths and negative health outcomes. The CDC estimates, for example, that second-hand smoke exposure results in 34,000 excess cardiac deaths each year.

Increased recognition of the health risks of passive smoke exposure lent significant political weight to anti-smoking efforts, resulting in a cultural shift over the last 30 years. As a result smoking has largely been banned in most indoor public places and many work places.

The empirical question on which Grier focuses is the impact of those smoking bans on health outcomes. He does a fairly thorough review of the literature, although I think his review is biased to make his point, that the health benefits of such bans have been overplayed and maybe don’t exist.

Conspicuously absent from his review is a 2016 Cochrane systematic review (the gold standard for systematic reviews in evidence-based medicine). This updated review concludes:

“Since the first version of this review was published, the current evidence provides more robust support for the previous conclusions that the introduction of a legislative smoking ban does lead to improved health outcomes through reduction in SHS for countries and their populations. The clearest evidence is observed in reduced admissions for acute coronary syndrome. There is evidence of reduced mortality from smoking-related illnesses at a national level. There is inconsistent evidence of an impact on respiratory and perinatal health outcomes, and on smoking prevalence and tobacco consumption.”

Another 2016 Cochrane review shows that institutional smoking bans decrease tobacco use and second-hand smoke exposure in hospital and universities, but not in prisons.

A 2017 systematic review concludes that public smoking bans do not shift smoking to the home, and does result in overall reduction in smoking.

A 2016 review focusing on child health also finds benefit:

“There is an increasing body of evidence that shows that children also experience health benefits after implementation of smoke-free legislation. In addition to protecting children from tobacco smoke in public, the link between smoke-free legislation and improved child health is likely to be mediated via a decline in smoking during pregnancy and reduced exposure in the home environment. Recent studies have found that the implementation of smoke-free legislation is associated with a substantial decrease in the number of perinatal deaths, preterm births and hospital attendance for respiratory tract infections and asthma in children, although such benefits are not found in each study.”

These are the most recent reviews that I could find and I think they fairly represent the state of the research. Grier gives an overall different view of the research, focusing on the fact that effect sizes have decreased as studies have increased in size and rigor.

No one should be surprised at this, however. In fact, the research on the impact of smoking ban legislation is an excellent example of the general trend in scientific research that I frequently discuss here. Preliminary research has a massive false positive bias, partly researcher bias and partly publication bias. As follow up research is more rigorous and thorough, effect sizes tend to decrease (this is termed the decline effect). This is a fairly consistent effect in the literature, but can be made to seem sinister if presented in isolation regarding one topic.

Grier implies that the benefit of such bans declines to near zero, but I don’t think the most recent systematic reviews justify that conclusion. Initial reported benefits were unrealistically high (such as a 60% reduction in heart attacks) but that should not establish the bar for later comparison. We have to look at the benefits on their own.

Since we are talking about public health measures, we need to consider population-based reductions in negative health outcomes, and the reviews show they are real and substantial. Are they enough to justify the current bans (or even expanding the bans)? Now we are getting into ethical philosophy, which can be informed by the science but also requires some value judgments.

The Ethics of Smoking Bans

Grier is a libertarian who discloses:

“I worked at the Cato Institute almost a decade ago when it received some tobacco company donations. Also, as part of my career as a bartender, I made cocktails at a 2016 event sponsored by Diamond Crown; I wasn’t paid, but I was given a humidor and cigars as thanks.”

Take that for what it’s worth. This is not necessarily a criticism, it is just pointing out that he is coming from a certain perspective in terms of which values he may deem more important (i.e. personal liberty).

I am a physician, and I freely admit I am coming from a different perspective. To me smoking is the enemy and we should do everything we can to reduce its incidence. I also freely admit that I personally hate second hand smoke. I find it an assault on my own personal liberty.

When considering the ethics of smoking we have to recognize the cultural and historical context. Smoking is accepted because of its long embedded cultural history. Imagine if a product such as smoking were just developed, and a company tried to introduce it to the market. This product is demonstrably harmful when used as directed. It increases the risk of lung cancer, heart disease, strokes, and respiratory illness. It increases these same risks to a lesser degree in those passively exposed to it. The product is also addictive, which compromises the free choice of the user. Use of the product also has zero benefit – it is purely for recreation.

I think it is reasonable to conclude that such a product, if a company tried to introduce it today, would never get past regulation or become legal. Tobacco is essentially a legacy product that we tolerate because of its history.

In fact, there is a reasonable argument to be made for outright banning of tobacco as a harmful addictive product. There are two arguments in favor of not outright banning tobacco. The first is that people have the right to make such decisions for themselves. The second is that a ban would be impractical, and would just create a black market and an expensive regulatory burden.

These are reasonable arguments, but also consider the cost to society of smoking. There are also many precedents for laws that protect the public from their own bad decisions, such as seatbelt and helmet laws. We all collectively pay the health care costs of other people’s bad decisions.

Such considerations absolutely have to be weighed against the liberty cost of regulation. A hyperintrusive nanny state can also have unintended negative consequences.

To be clear, I am not advocating for an outright ban, because I don’t think it would be culturally acceptable or practical. Consenting adults should reasonably have the right to engage in risky behavior, as long as they are properly informed and their actions do not hurt others. They should be prepared to pay the cost of such behavior, however, such as higher insurance premiums.

Short of an outright ban, there are a number of reasonable regulations that can mitigate the negative health effects of smoking. Further, it is generally recognized by ethics philosophers that negative rights outweigh positive rights. Person A’s right not to have something done to them outweighs person B’s right to do something, all other things being equal.

What this means, in my opinion, is that a person’s right not to be exposed to smoke outweighs someone else’s right to smoke. No one should be forced to be exposed to smoke in order to take a flight, or eat in a public restaurant, or work at their job. If you make the personal choice to engage in an unhealthful recreation, then the burden should be on you to do it in such a way that you do not impose that choice on others.

Grier complains that smokers have been increasingly stigmatized. I would argue that smoking has been stigmatized, but not necessarily smokers. In any case, this is a good thing. Even if as a society we do not ban a voluntary unhealthful behavior, we can still recognize that it is a poor choice that should actively be discouraged. For decades smoking had positive PR, and it was shown to be cool and glamorous. Now the tide has shifted and it is portrayed as a weakness. That is a good thing, and one could argue a fair and reasonable fix for prior misleading advertising.

Grier acknowledges that public smoking bans are now culturally embedded and probably not going anywhere. This is a good thing. Now that non-smokers are used to living their lives without having second hand smoke imposed upon them, it will be difficult to go back. I see this as just a correction of the prior unfair cultural inertia of smoking. It was tolerated far too much for far too long simply because it was embedded in our culture.

Keep one more thing in mind – about 70% of smokers actually want to quit. That is a testament to the addictive quality of tobacco. If they want to quit, it is difficult to argue against legislation that may help them do so, or at least reduce their smoking. In that sense we haven’t gone far enough.

97 responses so far