Smoking bans at Slate, plus a new study from Japan

Longtime readers of my blog may remember the “heart miracle” studies of the previous decade. These studies purported to show that implementing smoking bans would bring about drastic and immediate reductions in the rate of heart attacks. I and several other writers expressed doubt for a variety of reasons, noting the small sample sizes and methodological oddities that seemed to point to a pre-determined conclusion. At the time, however, there weren’t many large scale studies being done that could settle the question.

Those large scale studies now exist, and as I write today at Slate, the evidence for the heart miracle hypothesis isn’t holding up:

[…] now that the evidence has had time to accumulate, it’s also become clear that the extravagant promises made by anti-smoking groups—that implementing bans would bring about extraordinary improvements in cardiac health—never materialized. Newer, better studies with much larger sample sizes have found little to no correlation between smoking bans and short-term incidence of heart attacks, and certainly nothing remotely close to the 60 percent reduction that was claimed in Helena. The updated science debunks the alarmist fantasies that were used to sell smoking bans to the public, allowing for a more sober analysis suggesting that current restrictions on smoking are extreme from a risk-reduction standpoint.

It’s a rather long piece and it covers a lot of ground, from the original heart miracle in Helena, Montana to the pervasive outdoor smoking bans that stigmatize smokers today. Read the whole thing.

During the editing of the article I also came across yet another new study from Japan. This one compares data from Hyogo Prefecture, one of the first regions in Japan to impose a smoking ban, to the control population of Gifu Prefecture. The time period is fairly long (one year before the ban, two years after) and the populations are large (5.58 million residents in Hyogo, 2.07 million in Gifu). One limitation of the study is that the Hyogo ban isn’t 100% comprehensive; businesses such as bars can allow smoking and some other businesses can have separate smoking rooms. Nonetheless, given the pervasiveness of smoking in Japan, even a partial ban would be expected to significantly reduce non-smokers’s exposure to secondhand smoke.

So what were the results? No trend in the number of acute coronary syndrome admissions appeared in either prefecture. “For the primary endpoint of this study, we did not observe a significant change from before to after the implementation of the partial smoking ban.”

The study did identify a small downward trend in Kobe City, which the authors attempt to spin as evidence that the smoking was in fact having the desired effect:

The reason why only Kobe City showed a significant decrease in the number of ACS admissions irrespective of subgroups is unclear. One possible reason is that the Hyogo Prefectural Capital Office is located in Kobe City, and social understanding of smoking legislation might have been accepted more widely. Indeed, questionnaires by Hyogo Prefectural Government Health & Welfare Department, distributed in the bars and restaurants larger than 100m2 in 2015, showed that the adherence rate to the smoking ban legislation was 97% in Kobe City and 88% in other Hyogo districts included in the present study. Therefore, compared with the other districts in Hyogo, the adherence rate to the smoking ban legislation was higher in Kobe City.

OK, maybe, but this seems like a stretch. One should always be skeptical of post-hoc attempts to explain why the effect one was looking for only appears in a certain subset of the data. It’s easy to come up with just-so stories that fit the expected narrative. There are myriad other factors that could be causing the decline, and chalking it up to slightly better adherence to an already partial smoking ban strikes me as a very unlikely (and very convenient!) candidate.

In any case, the bulk of high-quality research published in recent years weighs heavily against the idea that smoking bans will bring about miraculous health benefits. So as I argue in my Slate piece, let’s move on to a post-miraculous policy and make sensible accommodation for smokers’ preferences.


How the FDA will snuff out innovation in cigars and e-cigs

I awoke yesterday morning to the news I’ve been dreading for years: The FDA’s Center for Tobacco Products announced that it will extend its authority over cigarettes to all tobacco products, including cigars and e-cigarettes. When I last covered this topic, the FDA was considering a proposal (“Option 2”) to exempt premium cigars from some of the agency’s more onerous regulations. I was optimistic that the agency might take this path, allowing it to concentrate resources where they might do more good and avoiding conflict with politically vocal cigar smokers. Unfortunately, as I write at Reason, that optimism was misplaced:

As of yesterday, Option 2 is dead. And so, perhaps, is innovation in the cigar business.

Cigars that were on the market in 2007 will be allowed to remain for sale, but any cigars introduced since then will have to endure the same sort of regulatory hassles as Hestia tobacco. If they can’t prove they’re substantially equivalent to existing products—not just in their composition or their effects on smokers, but in their essentially unknowable potential health impacts on the population as a whole—then they will be ordered off the market.

It’s hard to predict what those applications will cost, but the most likely outcome is that the market for cigars will soon become a lot less diverse and a lot more boring. (Cuban cigars, which by definition were not legally on the US market in 2007, will obviously not be grandfathered in.)

Read the rest here.


The blend that dare not speak its name

Today at Reason, I write about antiquated liquor laws that forbid producers from being completely honest with consumers:

Like many other whisky brands, Compass Box doesn’t distill their own spirits. They source whisky from other producers to create unique, proprietary blends. And like most companies making blended whisky, they tend to keep their precise recipes secret.

But for these two blends they took the unusual step of posting infographics on their website that provided detailed breakdowns of every component whisky, including the source distillery, tasting notes, the exact proportion each takes up in the blend, the type of cask used for ageing, and the length of time each whisky spent in barrel.

For the type of whisky drinker who’s willing to shell out three figures for an exclusive bottle, Compass Box’s complete transparency is a welcome departure from brands that obscure the provenance of their spirits with varying degrees of honesty. But at least one competitor viewed Compass Box’s openness as a violation of liquor marketing regulations. An anonymous distiller contacted the Scotch Whisky Association, a trade group for Scotch, who in turn informed Compass Box that its detailed disclosure was illegal.

The article covers not just whisky in the European Union, but also gin and aquavit in the United States. Read the whole thing.

I was also quoted in The Wall Street Journal this week in an article about reviving hot drinks made with beer. Readers interested in trying these drinks at home should pick up my book, which has an entire chapter dedicated to “hot helpers.”


The case for home smoking bans, Voxsplained

This one headline perfectly encapsulates why’s coverage of tobacco policy is terrible:


That’s a lie, actually. Vox’s tobacco coverage is bad for more reasons than can be encapsulated in one headline, and it’s not really much better or worse than any other publication’s, but I’m trying to keep with the current form of writing on the internet. “Not that the old way was perfect,” writes my old school blog pal Jason Kuznicki, “but nobody fisks anymore, and for that we ought to be ashamed.” I agree. So how about some fisking? Perhaps my rank on Technorati will go up if you link to this.

Here’s Vox’s German Lopez writing on the Department of Housing Urban Development’s announcement that it will be banning smoking in all 1.2 million of the nation’s public housing units. “One major problem with this policy is it seems to single out low-income people,” writes Lopez, providing a glimmer of hope that he might think twice about intruding into 1.2 million homes of the less well off. “But there’s an easy solution to that: Indoor smoking should be banned everywhere — inside bars, restaurants, your home. Full stop.”

Never mind the casual authoritarianism behind the sentiment, or the constitutional issues it raises. (Has the Commerce Clause been interpreted so broadly as to allow this sort of thing? Eh, probably.) And never mind whether enforcing this particular ban by evicting some of the poorest members of society from their homes is a humane idea — Megan McArdle covered that here. Let’s just look at some numbers, the stock in trade of explanatory journalism. Lopez supports his advocacy for enforcing a smoking ban even in privately owned homes by noting the scope of the problem:

According to the Centers for Disease Control and Prevention’s most recent data, smoking kills 480,000 people each year. Secondhand smoke alone kills nearly 42,000 people. To put that in perspective, that’s around 8,000 more people than die to either car crashes or gun violence.

The 480,000 deaths figure got picked up in just about every story about the smoking ban in federal housing, including additional Vox posts this weekend from Julia Belluz and Dylan Matthews, from whom I’d expect better. (Matthews also seems to wish that we could ban smoking in private homes, but regrets that people would “freak out” if we tried. Imagine!) Both the total number of deaths and the number attributable to secondhand smoke are dubious, however, and none of the writers appear to have put any effort into understanding where they come from.

Let’s take the total number first. As Belluz notes, the smoking rate in the United States has dropped from 21 percent of Americans in 2005 to 17 percent in 2014. The decline gets more significant the farther back you look. Yet the CDC’s estimates of mortality caused by smoking don’t seem to reflect this. The CDC currently tells the public that smoking kills 480,000 people per year. That’s nearly 40,000 more annual deaths than it attributed to smoking from 2000-2004, when it claimed that smoking killed 443,000 people. And that’s more than the 438,000 deaths per year the CDC assigned to smoking from 1997-2001.

It’s worth asking why smoking mortality seems be increasing even as smoking rates are doing down. This is partially due to population growth and demographic changes; mortality rates would be a far more useful metric for comparison over time. And since smoking is associated with chronic diseases, some lag in the figures is to be expected. But still, the 480,000 figure is controversial.

Unlike deaths due to car accidents and gun shots, which are discrete events that can be tallied, there’s no direct way to count deaths caused by smoking (or especially deaths caused by secondhand smoke). Death certificates don’t say things like, “Bob died from heart disease made marginally worse by his cigarette habit.” Instead, researchers compare the prevalence of causes of death that are associated with smoking to the rate of smoking among different age groups, attempt to account for other confounding factors, and give their best shot at an estimate of how many people who died would still be alive in a tobacco-free world. Exactly how the current guess of 480,000 deaths is arrived at isn’t really clear.

The CDC’s numbers have been criticized in academic journals. A 2007 paper by Brad Rodu and Philip Cole in Nicotine and Tobacco Research offers a different model, which Rodu summarized in an article for Cato’s Regulation that criticizes the lack of transparency in the CDC estimates. With the Rodu-Cole model, you see the good news you’d expect from fifty years of declining smoking rates: “The U.S. mortality rate attributable to smoking declined about 35% between 1987 and 2002.”

OK, that’s just one paper, and from a source you may consider less reliable than the CDC. But in 2012 the same journal published another paper by Brian Rostron, whose affiliation is with the Food and Drug Administration’s Center for Tobacco Products. Rostron is also critical of the CDC’s methods, noting that they “have not been substantially revised since their introduction in the 1980s.” Rostron’s revised estimate for annual smoking-related deaths in 2004 is 380,000. The paper concludes, “we have estimated smoking-attributable mortality over time and found that the number of estimated deaths has peaked and finally begun to decline for both men and women in the United States.” If that’s correct, then we should have experienced a decade of declining deaths since 2004, and the CDC’s 480,000 figure is wildly off target.

[Update: German Lopez brings up a new paper in the New England Journal of Medicine, which suggests some excess mortality missed by earlier figures due to diseases that have not yet been definitively linked to smoking. Some of the relative risks are low, and there may be confounding variables, but the sample size is large. If those links bear out, then the sets of estimates above would be revised upward.]

That’s the total number of deaths attributable to smoking. How about the fraction attributed to secondhand smoke? If it’s difficult to figure out how many smokers are dying because of tobacco use every year, guessing how many people die from secondhand smoke is even more problematic. The CDC currently pegs the number at 41,000 deaths per year, of which a little over 7,000 are due to lung cancer and about 34,000 are due to heart disease.

Let’s tackle lung cancer first. If you want a rigorous, scientific indictment of secondhand smoke as a cause of lung cancer, you probably can’t do better than the 2006 report from The Surgeon General. The report concludes unequivocally that environmental tobacco smoke increases the risk of lung cancer in non-smokers. But by how much? This is expressed as a relative risk. A relative risk of 1 would indicate that people exposed to secondhand smoke are no more likely to develop lung cancer than those who are not exposed. (People who smoke habitually have a relative risk in the neighborhood of 16, which is huge.) What would you guess is the relative risk for non-smokers who are exposed to secondhand smoke over the long-term at their home or workplace?

If you guessed barely greater than 1, you’re correct. The report’s table of meta-analyses puts the relative risks of exposure at home for non-smoking spouses or at work for non-smoking employees in the range of 1.12 at the low end to 1.43 on the high end. That’s really low! Low enough to be sensitive to bias in how various studies are weighted, and low enough that it can’t be reliably detected in studies of secondhand smoke exposure. (One of the meta-analyses of childhood secondhand smoke exposure even suggests that children who are exposed to tobacco smoke are less likely to develop lung cancer. As they say on Twitter, “Whoa, if true!”)

How you interpret this ambiguity likely depends on your political priors. With that in mind it’s interesting to see the Journal of the National Cancer Institute publishing a headline like “No Clear Link Between Passive Smoking and Lung Cancer.” That’s a 2013 story about a cohort study of 76,000 women that “confirmed a strong association between cigarette smoking and lung cancer but found no link between the disease and secondhand smoke.” To go on:

The incidence of lung cancer was 13 times higher in current smokers and four times higher in former smokers than in never-smokers, and the relationship for both current and former smokers depended on level of exposure. However, among women who had never smoked, exposure to passive smoking overall, and to most categories of passive smoking, did not statistically significantly increase lung cancer risk. The only category of exposure that showed a trend toward increased risk was living in the same house with a smoker for 30 years or more. In that group, the hazard ratio for developing lung cancer was 1.61, but the confidence interval included 1.00, making the finding of only borderline statistical significance. […]

But many studies that showed the strongest links between secondhand smoke and lung cancer were case–control studies, which can suffer from recall bias: People who develop a disease that might be related to passive smoking are more likely to recall being exposed to passive smoking. […]

However, Silvestri finds some reassurance in the passive-smoking findings. “We can never predict who is going to develop lung cancer,” he said. “There are other modifiers. But you can say, with regard to passive smoke, it’s only the heaviest exposure that produces the risk. We kind of knew that before, but it’s a little stronger here.”

“We’ve gotten smoking out of bars and restaurants on the basis of the fact that you and I and other nonsmokers don’t want to die,” said Silvestri. “The reality is, we probably won’t.”

The study has not, to my knowledge, appeared in a journal, but if you’ve followed debates about secondhand smoke at all, then the candor in the commentary above is refreshing. The association between secondhand smoke and lung cancer has never been as scary as anti-smoking advocates make it out to be. Read Christopher Snowdon for a sense of the numbers; the ten studies with the largest sample sizes find risk ratios of 1.29, 1.11, .70, 1.03, 1.53, 1.10, .90, and .96. This is hardly the stuff of nightmares, and it’s difficult to imagine basing an accurate model of secondhand smoke mortality on such tiny risks.

How about deaths due to heart disease? Economist Kip Viscusi provides what I think is a fair summary in his 2002 book Smoke-Filled Rooms: “Despite the central role of lung cancer in the ETS [environmental tobacco smoke] debates, the heart disease linkage may be greater, and the risk levels much larger.” This accords with the CDC’s higher attribution of deaths due to heart disease than to lung cancer.

Studies on the short-term effect of smoking bans, however, are not encouraging. In the early 2000s, a wave of research conducted in small cities that had implemented smoking bans concluded that they could drastically reduce heart attacks. Skeptics criticized these studies for relying on data from small populations. Newer research confirms that they were right to be doubtful. The most notable example is a 2014 study of the state of Colorado in The American Journal of Medicine which was co-authored by researchers who had previously published papers showing that small towns in Colorado experienced declines in heart attacks after banning smoking. In their new research, they write:

Although local smoking ordinances in Colorado previously suggested a reduction in acute myocardial infarction hospitalizations, no significant impact of smoke-free legislation was demonstrated at the state level, even after accounting for preexisting ordinances.


These analyses support the hypothesis that small study populations may be more likely to find dramatic changes in acute myocardial infarction incidence, whereas increasing the study sample size attenuates the magnitude of the reduction. Also, review of the studies in aggregate reveals data asymmetry that suggests the potential for publication bias or heterogeneity not entirely explained by a random-effects meta-analysis. […] Overall, a review of published research shows that acute myocardial infarction RR reduction appears inversely related to sample size. […]

Available evidence suggests that acute myocardial infarction incidence has been decreasing dramatically, unrelated to smoke-free ordinances. […] This emerging evidence highlights the importance of accounting for secular trends in acute myocardial infarction incidence before definitive attribution to smoke-free ordinances can be made. […]

Overall, available evidence suggests that the decrease in acute myocardial infarction incidence associated with reductions in secondhand smoke exposure may be substantially lower than originally estimated.

The decline in hospitalizations due to heart attacks is an important factor to consider. From the same paper:

Data from the Centers for Disease Control National Environmental Public Health Tracking Network recently evaluated secular trends in 20 Network states from 2000 to 2008 using a longitudinal linear mixed effects model. The authors documented a statistically significant overall decrease in age-adjusted acute myocardial infarction hospitalization rates, with most states showing more than a 20% decline during the period. This temporal reduction in acute myocardial infarction incidence is of a magnitude that exceeds the reduction observed in many smoking ordinance studies. Despite this, some analyses have not accounted for secular trends.

This reduction in heart attacks doesn’t get much attention, but it comes up in Rostrom’s paper as well: “Deaths from ischemic heart disease for persons aged 65 and over decreased from 201,000 in 2000 to 158,000 in 2007 for U.S. men, and from 233,000 in 2000 to 170,000 in 2007 for U.S. women.”

And you see similar downward trends in the UK. A study of Scotland from 2002-2010 found that deaths from heart disease declined by 43%. Nearly half of this decline was credited to improved treatment. Changes in the rate of smoking were credited for only 4% of the decline. One has to wonder: Given all the advances of the past decade or so, why is the CDC’s estimate of secondhand hand smoke deaths caused by heart disease still so high?

The decline in heart attacks and deaths due to heart disease has occurred over a similar time period as the spread of smoking bans. The best source for information on the adoption of smoking bans in the United States is Americans for Nonsmokers Rights, who maintains a database of smoking restrictions. According to their tracking, the number of jurisdictions in the United States with 100% smokefree laws in all workplaces, restaurants, and bars has increased from two in 1993 to 790 in October 2015. Estimates of the number of deaths caused by secondhand smoke, however, have been strangely consistent. The current estimate from the CDC is 41,000. That’s a little less than the 2006 Surgeon General’s report estimate of nearly 50,000. And if you go back to 1990, you can learn from The New York Times that:

The newer understanding of the health hazards of passive smoking were underscored in a report at a world conference on lung health in Boston last week. Dr. Stanton A. Glantz of the University of California at San Francisco estimated that passive smoke killed 50,000 Americans a year, two-thirds of whom died of heart disease.

Sound familiar? It’s almost as if the claim that secondhand smoke kills around 50,000 Americans per year is based on its political utility rather than any firm grounding in epidemiology.

To review: In the past two decades, the rate of smoking has substantially declined. Treatment and prevention of heart disease, which allegedly causes the vast majority of deaths associated with secondhand smoke, has improved dramatically. Doubts about the magnitude of the association between secondhand smoke and both lung cancer and heart attack incidence have spread to mainstream academic journals. And exposure to secondhand smoke has been greatly reduced thanks to declining smoking rates, the proliferation of smoking bans, and changing social norms. Yet the number of Americans dying from secondhand smoke exposure has, supposedly, barely nudged downward over twenty-five years. These facts don’t hang together.

So how many deaths does secondhand smoke cause per year? I’ve been writing about tobacco policy intermittently for nearly a decade and in-depth for the past year, and I’m reluctant to commit to a number. I can tell you that I’m extremely skeptical of the CDC’s figure of 41,000, and I think that any responsible journalist ought to be skeptical too.

This brings up problems with contemporary reporting on tobacco policy and with explanatory journalism in general. In the old days of tobacco reporting, a policy announcement such as the ban on smoking in public housing would have gotten just as much coverage as it did this week. But reporters covering it might have also sought comment from pro-smoking sources. A lot of what those sources said would have been total spin and bullshit of the sort satirized in Christopher Buckley’s Thank You for Smoking, but some of what they said might have been valid criticism that pointed writers to legitimate weaknesses in their stories. That dynamic has been much reduced since the Master Settlement Agreement of 1998 that dismantled pro-tobacco organizations.

A couple years ago I interviewed Michael Siegel, a professor at the Boston University School of Health who formerly worked for the CDC’s Office on Smoking and Health, and he was blunt about the deterioration of scientific integrity in the anti-smoking movement. Siegel is generally in favor of indoor smoking bans, but has become a vocal critic of how anti-tobacco groups’ exaggerate the dangers of secondhand smoke. He blames this in part on the loss of an adversarial process:

The current state of tobacco control I would describe, quite sadly, as misguided. It is now guided more by ideology and politics than by science. Ironically, I think one of the reasons it has lost its way is that some time around 2000 or so, the tobacco industry relinquished its watchdog role. Organizations in tobacco control used to be very careful because they knew the tobacco industry was watching and would call them on it if they exaggerated or distorted the truth. But after around 2000, the tobacco companies stopped playing this role and basically allow the tobacco control groups to say anything they want.

Good journalism is more than regurgitating a scary-sounding number from an authoritative source. It’s understanding the motivation and reliability of your sources, and seeking out potential opposition to see if there’s a counterargument that they failed to mention. Reporters know to be skeptical of tobacco companies. They haven’t learned to be skeptical of anti-smoking sources, and given the changes in the regulatory landscape of tobacco it’s long past time that they do. It’s the only way to avoid uncritically reporting that a smoking ban can decrease heart attacks by 60% in just six months, or that smokers are “contaminated” and “actually emit toxins,” or that people who use e-cigarettes are “inhaling Chinese-made antifreeze,” or that 41,000 Americans are dying from secondhand smoke exposure in 2015.

The past few days of coverage at Vox, with three different writers repeating the same dubious statistic and none of them investigating it, doesn’t raise my confidence in their model of explanatory journalism. This is explanation divorced from skepticism. The current media environment makes it easy to find numbers to support one’s political view with just a few minutes of Googling, appearing to write from a perspective of data-driven empiricism, and harvesting those all-important clicks. (Is Vox’s “The case for banning smoking indoors — even in your home” any less dumb than Slate’s recent piece arguing that spooning is sexist?) But when those numbers are taken at face value and without context, the writers are just one bad statistic away from calling in the cops to search for ash trays on your kitchen counter.

I like Vox. I really do, even if I find myself turning more often to their entertainment writing than their policy pieces. I view the site as a generally useful source of information about complex topics that it’s difficult for any one person to know in detail. Yet occasionally the site covers a topic that I do know in detail. And when it does, I have to wonder how much of their other coverage is equally superficial and credulous.


Smokers and health insurance: good and bad news

Looking through my web stats the other day, I noticed that some of the common searches bringing readers to my site relate to smoking and health insurance. Cigar smokers in particular seem to be interested in this. And for good reason: The Affordable Care Act generally requires health insurers to treat applicants equally, but it makes an exception for smokers. Premiums for smokers are allowed to rise up to 50% higher than for comparable non-smokers.

I haven’t written about this topic since early 2013, but a recent post at the Cato Institute blog caught my attention. Aaron Yelowitz gathered data on premiums across states and concludes that much of the additional cost charged to smokers does not reflect actual risk, and that the additional premiums paid by smokers vary widely from state to state. In states with a large surcharge for smokers, such as Wyoming, this represents a substantial hidden tax on smoking:

Let’s consider a 27-year-old who doesn’t receive subsidies but is mandated to purchase health insurance. If a non-smoker lived in Cheyenne, WY, he or she could purchase Blue Cross Blue Shield of Wyoming – BlueSelect Silver ValueTwo Plus Dental plan for $334 per month. This plan has a $3,000 deductible and an out-of-pocket maximum of $6,600. If the 27-year-old smoked, the same plan would be $417 per month, or 24.9% higher. For a pack-a-day smoker, this represents a $2.72 per-pack increase in expenditure due to Obamacare.

For cigarette smokers, who tend to have lower incomes, that’s enough to threaten the affordability of health insurance. Interestingly, tobacco companies and many anti-smoking groups joined in opposing this aspect of the law for that reason.

That hidden tax is the bad news for cigarette smokers. But what about cigar smokers, or anyone who smokes only on occasion? When I wrote about this in early 2013, Health and Human Services had not yet settled upon a definition for who counts as a smoker for health insurance purposes. Most of the proposals for defining this status, including those from anti-smoking groups, asked only when a person last used to tobacco. One proposal asked only “Have you used tobacco in the last twelve months?” and “Are you currently using tobacco products?” As I noted at the time, this ignored frequency of use and was potentially very problematic:

A casual cigar smoker would have to answer yes to both questions posed by America’s Health Insurance Plans. Twelve months is a long time! Should someone who enjoys an occasional cigar have to pay 50% (or more) higher on their insurance premiums, the same penalty faced by pack-a-day smokers?

A sensible definition would address not only recency of tobacco use, but also frequency within that time period (and possibly the form of tobacco used).

Using the Wyoming example above, under this definition the annual hidden tax on a resident who smoked just one cigar per year would be nearly $1,000! Such an enormous penalty would clearly exceed any plausible estimate of risk.

Now for the good news. The final definition of who counts as a smoker (for plans that qualify under ACA) does take frequency of smoking into account. From the Federal Register:

In this final rule, we establish a definition of “tobacco use” that is based on the National Health Interview Survey, while setting forth the meaning of “some days” to ensure clarity for issuers and consumers. Specifically, for purposes of this final rule, we define “tobacco use” as use of tobacco on average of four or more times per week within no longer than the past six months

That’s great news for people who enjoy only an occasional cigar, who won’t have to face exorbitant penalties for low-risk behavior.

(This rule was published more than two years ago, by the way. I just now followed up on it, which says a lot about my commitment to blogging as a medium these days. But given that people find my site searching for this topic, it seemed worthy of a belated update.)


Gasp! New Orleans passes a smoking ban

In my first contribution to Eater, I put the New Orleans smoking ban into a national context:

As in most cities, New Orleans’ smoking ban debate centered on the trade-offs between the interests of business owners who fear lost patronage and the interests of customers and employees in avoiding secondhand smoke. It’s an apt time to inquire into what’s really at stake with smoking bans. Will the sky fall for bars and casinos in New Orleans? And will banning smoking save as many lives as advocates promise?

The answers may surprise you!


Vaping in the Oregonian

In today’s Oregonian, I look back at a column I wrote in 2008 and say, “I told you so.”

When the clock struck midnight on New Year’s Day 2009, Oregon ushered in its statewide smoking ban in bars and restaurants. I was at the Horse Brass Pub, one of Portland’s most notoriously smoky drinking dens, enjoying one last cigar with a bunch of other patrons who were none too happy about the new rules.

The ban, we were told, was necessary to protect employees and customers alike from secondhand smoke. Health researchers had conducted dozens of studies attempting to show that exposure endangered nonsmokers. Some of the results were medically implausible, but ban advocates at least made the effort of demonstrating actual harm to actual humans.

Many of us doubted that the evidence really mattered. As I wrote in The Oregonian/OregonLive at the time, “Protecting workers is simply the polite fiction by which nonsmokers have imposed their will on an increasingly unpopular minority.”

We suspected this, but how could we prove it? What if there were a device that looked like a cigarette and mimicked the effects of smoking, yet emitted a mostly harmless vapor instead of tobacco smoke? If authorities tried to ban that too, without bothering to establish that it endangered anyone, then our suspicions would be vindicated.

That device exists. It’s called an e-cigarette. And sure enough, the Multnomah Board of County Commissioners is voting on whether to ban its use indoors. The Legislature, too, may expand the state’s smoking ban to cover vaping.

Read the whole thing.


Dirty hands and raw meat

There’s nothing the media loves more than an elected Republican saying something stupid, and Republicans this week have been happy to oblige them. Rand Paul has, thankfully, backed off his anti-vaccine remarks, but Senator Thom Tillis stepped in to fill the void. He casually remarked that restaurants perhaps shouldn’t be legally forced to require employees to wash their hands, so long as they post signage alerting customers to the policy. Any restaurants doing so would presumably go quickly out of business. The free market at work!

This non-story was the most read political item at The Washington Post, it’s been covered by tons of news outlets, and my Facebook and Twitter feeds are full of people piling on in mockery of this call for extreme laissez faire capitalism.

But the thing is, Tillis was sort of right. Not about this particular regulation, one that has precisely zero constituency calling for its removal, but about the idea that consumers often ought to be allowed to opt-out of restrictive health codes to eat and drink what they like.

Health codes governing restaurants exist to protect consumers from risks that they can’t easily ascertain by themselves. As a diner, there’s no easy way to know whether the back of house is clean, whether the food has been stored safely, or whether contamination is occurring. Instead we have codes that govern these things and inspectors that pop in periodically to make sure everything is at it should be. It’s not a perfect system, but it works tolerably well.

But while codes are uniform, consumers’ risks preferences aren’t. Often health officials have concerns about an ingredient or cooking technique, but some subset of consumers demand it anyway. In those cases, informing consumers of the risk before they proceed is a viable compromise. This happens in restaurants all the time and it arguably should happen even more.

The most obvious example is the cooking temperature of meat. Health officials are wary of serving meat, especially hamburgers, that has not reached temperatures sufficient for killing E. coli. They advise against eating burgers cooked rare or medium rare. Consumers who know the pleasure of a perfectly cooked patty are willing to take the risk. Fortunately, under federal codes they have a choice. Restaurants can serve meat at lower temperatures as long as they put a warning on the menu along the lines of “consuming raw or undercooked meats may increase your risk of foodborne illness.”

In principle, that’s not too far off from what Tillis was suggesting. Even in 2014, undercooked meat isn’t legal everywhere. North Carolina didn’t legalize rare burgers until 2012, and I’m unclear on the law in South Carolina. All of Canada is a apparently a hamburger wasteland unless you find a chef willing to take his chances with the law:

The official “safe” temperature for hamburger meat, as enshrined in municipal codes and provincial acts across Canada, is 71 degrees Celsius, eight degrees higher than the generally accepted threshold for medium rare. […] “Every so often a restaurant will come up and advertise pink burgers, and we will go in and talk to them,” said Anna Marie D’Angelo, spokeswoman for Vancouver Coastal Health, the city’s health inspection agency. If West Coast diners see even a tinge of pink, health authorities advise them to “ask it to be recooked for their own safety,” said Ms. D’Angelo. Health Canada takes it a step further: After sending back the offending burger, “ask for a new bun and a clean plate, too” reads an advisory on the federal agency’s website.

These are real people with real jobs backed by real government power! They make chefs wary of speaking on the record about something as mundane as cooking hamburgers.

“I’ve served probably 100,000 burgers and nothing’s happened,” said Greg, a Canadian restaurant owner who isn’t in fact named Greg but wished to stay anonymous, arguing that media attention could attract unwanted scrutiny from the health department. Greg sources his own meat and grinds it in-house, but he still treads a narrow legal line. “A lot of guys do it, but we do it under the radar. If we put our names out there … they’re going to stop it.”

As with meat in the United States, so with sushi, oysters, eggs, and many other foods that are often best prepared raw or with minimal cooking. Want to dip your fries in aoili? Want to drink a cocktail shaken with egg whites? Want to sip Scotch in one of the few remaining smoking lounges? Then be glad health officials deign to let us off with just a warning.

In the United States, raw dairy is probably the best example of a food we should be able to eat given a clearly stated warning, but currently cannot. As cheese lovers know, the FDA forbids the interstate sale of raw milk cheeses that have been aged for fewer than sixty days, keeping delicious fresh cheeses off the market. Some states allow the sale of raw milk with a suitable advisory, while in others it’s completely banned, forcing consumers to buy on the black market or find legal workarounds. Thus we have the spectacle of federal agents conducting sting operations and arresting farmers for selling milk and cheese. Wouldn’t it better to let people make their own decisions after being suitably informed of the risks?

If you think so, then you and Thom Tillis are sort of on the same page. If you enjoy sushi, “undercooked” meat, various egg-based sauces and dishes, or lighting up in a cigar room, then you’re on board with the principle Tillis was clumsily attempting to illustrate. If you’ve experienced a perfectly cooked steak pulled from a sous vide bath, be glad New York officials stopped fining the chefs who use the technique and worked with them instead to write regulations that effectively govern it.

If we didn’t push back against risk-averse health officials about what we eat, drink, and smoke, our bars and restaurants would be a dull culinary landscape of overcooked food and excessively sanitized interiors. Everyone’s laughing at Tillis’ hypothetical hand-washing scenario, but it was just a year ago that California was stupidly ordering every bartender in the state to wear disposable latex gloves. Of all the people saying Tillis is nuts, foodies especially ought to know better. (I’m looking a you, Eater.)

As an illustration of excessive regulation, requiring employees to wash their hands after using the bathroom is obviously a dumb one. Tillis deserves to be mocked not because he’s wrong about regulation, but because he picked such a stupid way to make his point. That this is the best example he could come up with is, as Thoreau notes, just the latest evidence of the GOP’s utter uselessness.

But the gist of the argument isn’t crazy. There is, thankfully, no consumer demand for cooks with filthy hands and this isn’t a legal change that we need to be remotely worried about. We do, however, have reasons enough to be concerned about overzealous regulators interfering with the choices consenting adults make about food and drink. The principle at stake is worth defending, and it’s unfortunate that it falls to people like Tillis do it so ineptly.

[Photo used under Creative Commons license from roboppy.]


Break out the cigars? Not yet.

[Update 12/21/14: An expanded, more detailed version of this post is now up at Reason.]

Today’s announcement of more normal diplomatic relations between the United States and Cuba is welcome news for many reasons, but the one that seems to spring to mind for many is that Cuba’s most famous export, cigars, may finally become legally available in the United States. If only it were that simple. Here are three obstacles to getting Cuban tobacco into the US.

1. The embargo is still in place.

The new rules allow American travelers to return with up $400 of Cuban goods, of which only $100 worth can be alcohol or tobacco for personal use. This is still a long way from allowing commercial import. As before, any significant trade in Cuban cigars will be on the black market.

2. Trademark battles are going to be complicated.

The United States is the world’s largest market for premium cigars and our embargo with Cuba has essentially divided the global market in two: us and everybody else. Our embargo has created dueling trademarks for cigars. Cuban brands such as Cohiba, Partagas, Hoya de Monterey, Bolivar, and Punch are sold around the world. The United States does not recognize the Cuban trademarks, so cigars of non-Cuban origin with identical brand names are sold here. A legal dispute over the American and Cuban claims on the Cohiba trademark has dragged on since 1997 and still has not been resolved.

So even if the embargo is lifted entirely, the transition to allowing Cuban cigars won’t be a smooth one. Many of the Cuban cigars one can currently buy abroad would violate trademarks if imported to the US. Companies will have to fight this out in court or come to mutually beneficial agreements. Other likely outcomes are that Cuban products will be marketed under different names in the US or that non-Cuban companies will start using exported Cuban tobacco in their products. In any case, buying that Cuban Partagas Lusitania I enjoyed so much won’t be as simple as stopping into the nearest cigar store, at least in the short term.

3. The FDA could ruin everything.

The FDA has already announced that it intends to begin regulating cigars and outlined two approaches to doing this. The agency’s so-called “Option 2” would create an exemption for premium cigars, allowing them to be sold under less scrutiny than cigarettes and other tobacco products. The proposed standards for this exemption are problematic (see my Daily Beast article for details), but they would leave the door open for eventual Cuban imports.

The FDA’s Option 1, however, would be very bad news for Cuban cigars. Option 1 treats cigars just like cigarettes. Under the Tobacco Control Act, all tobacco products introduced to the United States after February 15, 2007, must receive explicit approval by the FDA. Getting approval is virtually impossible. As of my last coverage of the topic, only two new cigarettes had ever made it through the process, while thousands of product applications continue to languish in bureaucratic limbo. (See my articles in The Atlantic and Reason).

There were a lot of cigars legally on the market in 2007, but obviously none of them were Cuban. We don’t know yet know which option the FDA will choose, but Option 1 would have a disastrous impact on innovation in the cigar market. All Cuban imports and any new Cuban blends would have to somehow get past FDA regulators, whose record on cigarettes is terrible. The trade embargo could be lifted in the near future, just in time to have a new, de facto embargo imposed by bureaucrats at the FDA.

I’m keeping my fingers crossed for something like Option 2, but given political opposition to removing the embargo, complicated battles over trademarks, and byzantine regulations imposed by the FDA, I suspect that I’ll be slipping Cubans into my luggage on trips abroad for many years to come.

[Photo courtesy of Alex Brown on Flickr.]


Big Tobacco vs little e-cigs

Last month, the FDA closed its comment period for its proposed regulations on e-cigarettes, cigars, and other tobacco products. The three Big Tobacco companies are among the parties submitted comment. The submission from Reynolds was particularly self-serving:

Reynolds American Inc. has fired an expected shot across the bow of small vapor cigarette manufacturers.

A Reynolds division recommended to the Food and Drug Administration in a 119-page submission that the agency ban the use of vapor electronic cigarettes.

Traditional e-cigs are battery-powered devices that heat a liquid nicotine solution in a self-contained disposable cartridge and create a vapor that is inhaled. The manufacturers have provided few flavor choices, in part in expectations that the FDA would limit flavorings as they do with combustible cigarettes.

By comparison, vapor products can feature a liquid capsule that is inserted into a cartridge, known as an open-system format. Vapors offer consumers a wider variety of flavors, included fruits and candy.

The Big Tobacco companies have invested heavily in disposable cartridge e-cigs. These compete with the more customizable, DIY devices favored by many vapers and sold by small producers. But why try to win consumers over when you can persuade the government to ban the competition instead? The FDA’s new regulations give Big Tobacco the opportunity to try that, so of course they’re taking it.

This kind of regulatory capture has been baked into the FDA’s handling of tobacco from the beginning. The initial legislation was backed and negotiated by Philip Morris, owner of Marlboro; it benefited from new marketing restrictions and barriers to entry that protect its brands from competition. The market for cigarettes has been essentially frozen in its favor since the FDA’s extremely burdensome review process went into effect in 2009.

Reynolds is trying to repeat the same trick with e-cigarettes, stoking fears of an unregulated market to protect big brands from competition. They might succeed. Meanwhile the public benefits of FDA tobacco regulation remain extremely dubious.

Previously: My own comments to the FDA are here, and more detail on the FDA’s anti-competitive tobacco regulation is in my article for Reason.


Comment on the FDA’s cigar regulations

If you’ve been following my writing on the subject, you know that FDA regulation has the potential to devastate the market for cigars, pipes, and e-cigarettes. Tomorrow is the final day to submit public comments on the agency’s proposal to extend their authority to these products. Comments can be submitted here. Helpful suggestions for commenting can be found here from the site Halfwheel. My own comment, limited to cigars, is below.

Comments submitted for FDA regulations regarding premium cigars, Docket No. FDA-2014-N-0189:

As an avid, though only occasional consumer of premium cigars, I have been following closely the FDA’s regulation of tobacco products. The Tobacco Control Act empowered the FDA to review all new tobacco products before they come to market, with the aim of ensuring that these products are, at minimum, no more harmful to the health of society as a whole than the products already being sold in 2007. While this is arguably a laudable goal, in practice the implementation of pre-market review has been hampered by infeasible standards that render the market for tobacco less competitive without corresponding benefits to public health.

Though this consequence is unintended, it was not unforeseen. Testifying to Congress in 2007, then commissioner of the FDA Andrew C. von Eschenbach predicted that the law would not allow enough sufficient time for the agency to develop science-based rules regarding tobacco and would “unduly and unfairly raise the public’s expectations about what the Agency could accomplish.”

The agency’s record so far has shown that Eschenbach was correct. Since taking over regulation of cigarettes, the FDA has received nearly 4,000 applications for substantial equivalence. Only a tiny percentage of these have been approved, and these have mostly involved either very basic products (i.e. rolling papers) or very minor changes to existing products (i.e. substituting one type of cigarette paper for another). The vast majority of applications remain stuck in regulatory limbo.

The experience of companies trying to bring new products to market suggests that doing so is nearly impossible. Lorillard, one of the largest tobacco companies, was able to do so only after extensive delays going well beyond the 90 or 180 day deadlines implied by the Tobacco Control Act. The smaller startup Hestia has found itself mired in insurmountable bureaucracy. Documents provided by Hestia show that more than two years into the review of its substantial equivalence application, the FDA has not even begun to examine the physical characteristics of its cigarettes, focusing instead on marketing materials relating to identification of the predicate product. These details are irrelevant to the health impact of Hestia’s cigarettes and call into question the scientific basis of pre-market review. (Since pending applications are not made public, the experience of the vast majority of applicants remains unknown.)

The unintended consequence of this lengthy review process has been to freeze the market for cigarettes as it was in 2009, protecting the brands that dominated then from competition. The health benefits of restricting the entrance of competitors are unclear, especially when reviews do not reach the stage of evaluating the actual physical characteristics of new products.

Given that the market for premium cigars is much more dynamic than the market for cigarettes, with potentially thousands of new products being introduced each year, applying the FDA’s current rules and procedures to cigars would be devastating both to producers and to cigar smokers who value variety. It would also overwhelm the agency with applications, assuming producers bother attempting to navigate the review process.

This raises the question of how the FDA can regulate cigars without unduly burdening manufacturers. One possibility is to simply not apply the deeming regulation to cigars. This is my preferred course of action, though it is obviously unlikely that the FDA will accept it.

A second option is that proposed by the agency to exempt premium cigars from many of the regulations that apply to cigarettes (“Option 2” in the FDA’s proposal). This requires the creation of a legal distinction between premium and non-premium cigars. The agency’s proposal suggests eight factors that would distinguish premium cigars. They would be wrapped in whole tobacco leaf, contain 100% leaf tobacco binder and primarily long filler tobacco, be made by hand, lack any filter or mouthpiece, have no characterizing flavor other than tobacco, weigh no more than six pounds per 1,000 units, and retail for no less than $10 per cigar.

This definition shares many similarities with that in proposed legislation to exempt premium cigars from FDA regulation. The most glaring difference is the retail price requirement, which would set an effective price floor of $10 for all new cigars. This would be a very large price increase for consumers; one industry analysis finds that only about 15% of premium cigars currently sell for $10 or more.

The price of a cigar obviously has nothing to do with its objective characteristics that could affect a smoker’s health. A $5 cigar is no less or no more healthy than a $25 cigar. A $10 price floor would rightly be seen as politically expedient rather than scientifically justified, a means of keeping wealthy cigar smokers happy.

Another likely consequence of instituting a $10 price floor would be to incentivize black market sales. If all new premium cigars in the United States are required to retail for at least $10, consumers will be attracted to the greater variety and lower prices of cigars sold abroad (including those from Cuba, which are already illegal in the United States). Internet sales and in-person smuggling of foreign cigars would certainly increase. These black market cigars would not be regulated by the FDA at all, undermining the goals of regulation.

Nonetheless, the agency may conclude that some price floor is necessary to separate premium cigars from the cheap that allegedly draw in youth smokers. In that case, a compromise implementing a much lower price floor would better reflect the reality of the market.

A second objection to the proposed definition is the ban on all characterizing flavors other than tobacco. It is unclear what this would entail. Some flavors in cigars arise from ageing them in various woods. Many flavors such as whiskey or rum are not the kind one thinks of as being aimed at teenagers and are marketed to adults. A blanket ban on all flavors would be overly broad. Instead, the agency should work with the industry to identify types of flavoring that are of particular concern and evaluate them on a case-by-case basis.

A third objection is to the requirement that all premium cigars be produced entirely by hand. This also is unrelated to the health impact of cigars and would advantage producers in places with low labor costs. It would likely end some cigar production in the United States, which is often assisted by machine. This, too, is an area in which the agency could consult with the industry to refine its definition of premium cigars.

Finally, I would like to suggest broader changes to the way the FDA handles substantial equivalence applications. In current practice, applicants must identify a specific predicate product and provide detailed empirical analysis showing that their proposed new product is substantially identical to it, raising no new questions of public health. This is unduly burdensome on new producers and unfairly advantages existing players, who possess information about their own products and the funding to analyze them. Even when proposed new products raise no new questions of health, applicants may not have access to information about predicate products that would allow them to reach the empirical stage of FDA review. For example, applicants are expected to provide documents such as bills of lading from more than a decade ago to prove that their selected predicate product was marketed in 2007. If the owner of the predicate product is a potential competitor, they obviously have no incentive to assist in the provision of such materials.

To make the substantial equivalence application process more equitable, the agency should establish objective guidelines for new products that would allow them to be considered substantially equivalent to products already on the market. Current applicants do not know how much variation from predicate products is acceptable and often cannot access information about predicate products in the first place. Clear guidelines established by the FDA would give smaller producers a chance at navigating the process without sacrificing public health. The current substantial equivalence pathway would also remain open to those who prefer it.

Regardless of whether the substantial equivalence pathway is streamlined, I hope that the FDA will adopt some variation of Option 2, exempting premium cigars from most of the requirements of the Tobacco Control Act. The market for premium cigars is akin to those for craft beer, wine, or coffee, defined by skilled producers and specific origins of product, and completely unlike the relatively commodified market for cigarettes. It is incompatible with the expensive and time-consuming review currently required of new tobacco products. And as the FDA itself has noted in its proposal, requiring such review would likely accomplish little for public health, given that premium cigars are consumed and marketed to adults and that their use is less likely to lead to addiction. Existing research also suggests that moderate use of cigars is much less harmful to health than use of cigarettes.

The differences in the impact on population-level public health among individual cigars would be so small as to be impossible to ascertain. Given that the FDA’s Center for Tobacco Products has limited resources, its employees’ time can be put to much better use than the review of countless substantial equivalence applications for premium cigars.

– Jacob Grier
Portland, Oregon
August 7, 2014


Cigars at The Daily Beast

My first contribution to The Daily Beast explains what’s at stake in the proposed FDA cigar regulations:

Ever since the FDA was given authority over cigarettes in 2009, cigar makers have been pushing a bill in Congress to keep stogies out of the agency’s purview. That an industry would try to protect itself from FDA regulation is not surprising. That the FDA might agree with them is. And given the agency’s record on cigarettes, keeping its hands off of premium cigars is the right idea.

Read the whole thing here.


What proposed FDA regulations mean for e-cigarettes and cigars

I had a relaxing morning planned until the FDA announced its proposed regulations extending its authority to more tobacco products, including e-cigarettes and cigars. Predictably most of the press is focusing on the former, but the proposals over cigars are also very interesting. The long PDF detailing the proposal is here. Since you probably don’t want to read that, read this post instead.

First, a little background on how the existing law works with regard to cigarettes. The most important power that the Tobacco Control Act gave the FDA was pre-market review. Before releasing a new cigarette, producers must now get explicit approval from the FDA. This created three classes of products:

1. Products that were already on the market as of February 2007 are grandfathered in and allowed to be sold without review, although the FDA could hypothetically order them off the market.

2. Products introduced between February 2007 and March 2011 are allowed on the market while under provisional review. Producers have had to submit applications, but they are allowed to continue selling while the FDA reviews them.

3. Since March 2011, all new products must receive FDA approval before being sold.

This has resulted in a freeze of the cigarette market since 2011. In the three years since then, only two new cigarettes have been approved for sale. The anti-competitive effect this has had on the cigarette market has been my primary criticism of the FDA’s handling of tobacco. It has been reviewing applications for four years, currently has more than 150 employees working on reviews, has received approximately 4,000 applications, and has managed to rule on only 34 of them. For more detailed information on this, see my coverage in The Atlantic and Reason. (Keep in mind, too, that the Tobacco Control Act was fully backed and negotiated by Philip Morris. They knew what they were doing.)

It’s also helpful to know what the FDA is looking for in new product applications. The law establishes two routes to approval. One is for completely new products (premarket tobacco applications) and requires extremely burdensome amounts of data; this is basically uncharted territory at this point. The other is “substantial equivalence.” To get approval by this route, a new product must demonstrate that it is substantially equivalent to a predicate product that was already on the market as of February 2007 or has since been approved by the FDA. “Substantially equivalent” is defined to mean having the same characteristics (materials, ingredients, design, composition, heating source, or other features) or raising no new questions of health. (To see why this is a huge obstacle to new producers, see my articles above.)

It’s been known for a long time that the FDA planned to extend its authority beyond cigarettes. The biggest concern is how the agency’s sluggish review process will affect these new products, especially e-cigarettes and cigars. These are both dynamic sectors of the market and applying the same standards that the agency uses for relatively commodified cigarettes is extremely problematic.

Impact on e-cigarettes: As mentioned above, substantial equivalence applications must specify a predicate product by which to compare the new product. That predicate product must have been on the market by February 2007. You can see the problem here. The market for e-cigarettes barely existed then. Thus the review process as it exists now is essentially a death sentence for e-cigarettes. As the agency notes in its proposal today, its hands are tied: “Because this date is written into the statute, we do not believe that we have the authority to amend it with respect to e-cigarettes or other products.”

Because of this, the FDA’s proposal gives e-cigarette companies two years after the date the rule goes into effect to submit a premarket tobacco application (PMTA). What happens after that is anybody’s guess. But unless the law changes, it looks like the substantial equivalence option is off the table for e-cigarettes and the variety of products that remain for sale will be extremely restricted. If any products successfully navigate the PMTA process, they will likely be those with lots of financial backing and perhaps the right connections.

Impact on cigars: The FDA’s proposals regarding cigars are intriguing. The agency offered two options. Option 1 is to treat cigars just like other tobacco products, subjecting them to all the same burdens of review. Option 2 is to carve out an exemption for premium cigars.

The first option, as I’ve been warning for a long time, would be disastrous. Hundreds of new cigars come out every year in distinct blends, shapes, and ages. Forcing them into a review process that has managed to approve only two cigarettes in four years would destroy the market as we know it. It would also require all cigars to be substantially equivalent to those already on the market in 2007, making the sector considerably more boring.

The fact that Option 2 is even being considered shows that the FDA is aware of this. Under this option, exemptions would be made for premium cigars. A cigar would be exempt if it:

(1) Is wrapped in whole tobacco leaf

(2) contains a 100 percent leaf tobacco binder

(3) contains primarily long filler tobacco

(4) is made by combining manually the wrapper, filler, and binder

(5) has no filter, tip, or non-tobacco mouthpiece and is capped by

(6) has a retail price (after any discounts or coupons) of no less than $10 per cigar

(7) does not have a characterizing flavor other than tobacco**

(8) weighs more than 6 pounds per 1000 units.

This is not a perfect definition, but it’s a start. The biggest drawback is that it would create a price floor of $10 per stick and this price would be adjusted (i.e. increased) every two years. It’s still possible to get pretty nice cigars for under $10, especially if one buys them a box at a time, so this would be a substantial imposition on cigar smokers. At this point, however, I’m just glad that the option to exempt premium cigars from the FDA’s pre-market approval process exists at all.

Impact on pipes: Pipe smokers, now few and far between, did not put a substantial lobbying effort into influencing the FDA’s new regulations. As a result, pipes get very little discussion in the proposal. However pipe tobacco will be subject to the full authority of the FDA and, if I am reading it correctly, pipes themselves would be subject to pre-market review as well. That seems potentially problematic for unique, handmade briar and meerschaum pipes. Pipe smokers, stock up now or prepare to order from overseas!

What’s next: There are seventy-five days to comment on the proposed regulations. The biggest fight will be over e-cigarettes, which are the hottest topic in the press. In my view it’s a mistake for the FDA to wade into this until it gets its review process under control or can provide a workable alternative to the substantial equivalence path. The upside is that it will take at least two years before it takes enforcement action, which will allow more studies on e-cigarette’s effectiveness as a harm reduction tool to be conducted.

(Note also that the law requires the FDA to examine health impacts on the population level, not on the individual user, so it could order e-cigarettes off the market even though they are unambiguously safer than cigarettes. Remember too that the head of the FDA’s Center for Tobacco Products, Mitch Zeller, came directly to the job from consulting for GlaxoSmithKline, which makes nicotine replacement therapies that compete directly with e-cigs. Might pharmaceutical companies use the new restrictive review process to develop nicotine vapor devices of their own? I would not be at all surprised.)

Cigars have been almost completely off the radar of press and anti-smoking groups, but expect that to change as the debate over exempting them unfolds. Cigar smokers will have to continue keeping the pressure on lawmakers and the FDA to not destroy the industry. It will be important to show that premium cigars are primarily enjoyed by adults and have different health effects than cigarettes for the typical user. (For a summary of the latter, see here.)

Regardless of which option the FDA takes, I expect black market sales of cigars to increase. If it takes Option 1, the variety of cigars available in the United States will suffer greatly. If it takes Option 2, the price will rise to $10 a stogie.* Cuban cigars are pretty alluring already! If you enjoy the company of your local tobacconist, savor the next few years you have together. Their store may not be around much longer.

* Update to add that cigars could escape the price floor by winning FDA approval, but given the agency’s record so far I would not expect many to achieve that. Cigars that were on the market before 2007 could remain available at a lower price too.

** Additional note: I’ve asked the FDA several times whether ageing tobacco in cedar wood, a traditional practice for many cigars, would run afoul of the rule against characterizing flavors. They have not been willing to clarify this.


Multnomah County misleads bar owners

As a follow-up to my article in yesterday’s Oregonian about the failed attempt to include e-cigarettes in the state’s smoking ban, today I’m posting a memo Multnomah County officials have sent to local bars and restaurants. In it, they mislead business owners about the dangers of e-cigarettes, telling them:

State law does not currently prevent the use of e-cigs; however business owners are encouraged to include e-cigs in no-smoking policies. E-cigs pose serious health risks and challenges to enforcement of the Smokefree Workplace Law as it appears people are smoking indoors.

The letter then recommends that businesses include e-cigarettes in their no-smoking policies, adopt completely smokefree outdoor dining areas, and adopt a completely tobacco-free policy for their entire properties. (Here’s a PDF of the memo.)

There are valid reasons why a bar or restaurants might ban the use of e-cigs, such as the fact that some guests find them annoying. But county officials’ claim that the devices pose “serious health risks” is completely unsubstantiated. There’s not even much evidence that e-cigarettes are dangerous for users, much less for bystanders exposed to vapor secondhand.

There have been two recent studies on exposure to e-cigarettes in realistic indoor conditions. They are summarized here and here. Conclusion of the first:

… the quality and quantity of chemicals released in the environment [by vaping] are by far less harmful for the human health compared to regular tobacco cigarettes. Evaporation instead of burning, absence of several harmful chemicals from the liquids and absence of sidestream smoking from the use of the e-CIG are probable reasons for the difference in results.

And the second:

The study showed that e-cigarettes are a source of secondhand exposure to nicotine but not to combustion toxicants… Using an e-cigarette in indoor environments may involuntarily expose nonusers to nicotine but not to toxic tobacco-specific combustion products.

Even in the case of nicotine, exposure from real cigarettes was ten times higher than that from e-cigarettes.

Those are studies of indoor use. Multnomah County’s advice is to ban them outdoors too. The idea that indoor e-cigarette use could be harmful to bystanders is at least worthy of investigation, although the evidence so far is that it’s nothing to worry about. The idea that outdoor use presents serious health risks is wildly implausible.

This is yet another example of how the crusade against e-cigarettes is driven by unscientific alarmism rather than any empirical evidence of danger. County officials have shown that they have no credibility on the issue by misleading local business owners about the alleged risks.


Oregon’s war on vaping

The Oregon legislature recently failed to pass completely sensible restrictions on selling e-cigarettes to minor, an effort undermined by more extreme anti-smokers who were more intent on banning vaping in workplaces, bars, and restaurants. In today’s Oregonian, I write about lawmakers’ misguided attempt to include e-cigarettes in the smoking ban and their next proposal to impose new taxes on them. An excerpt:

I never thought I’d say this, but I’m actually getting nostalgic for the original smoking ban debate. Advocates exaggerated the dangers of secondhand smoke, but at least they made an effort to ground their views in science and demonstrate that non-smokers were being harmed.

The same cannot be said for those seeking to extend current bans to cover vaping. They’ll be the first to tell you that more study of e-cigarettes is needed. But why wait for results? They’re ready to ban first and ask questions later.

Read the whole thing here.


New at Reason

After a long absence (my previous article was in 2008!), I’m back in the pages of one of my favorite publications today. Over at Reason I take a look at the FDA’s latest actions against tobacco, explain why they accomplish nothing for public health, and spell out what they imply for the future of cigars and e-cigarettes.

The article also updates the case of Hestia Tobacco, whose regulatory tribulations I profiled at The Atlantic one year ago. Unfortunately, they are no closer to coming to market than they were before. See that article for a more in-depth explanation of the laws that allow the FDA to keep new tobacco products in regulatory limbo.

Finally, be sure to check in at Michael Siegel’s blog, where he has been tearing apart the FDA’s action this week from a slightly different perspective.


How grilling meat really is like a smoking a cigar

In his “Explainer” column at Slate yesterday, Brian Palmer raised alarm about grilled and smoked meats, suggesting that by eating them we may be, as the headline puts it, “Cooking Up Cancer”:

A growing body of research suggests that cooking meats over a flame is linked to cancer. Combusting wood, gas, or charcoal emits chemicals known as polycyclic aromatic hydrocarbons. Exposure to these so-called PAHs is known to cause skin, liver, stomach, and several other types of cancer in lab animals. Epidemiological studies link occupational exposure to PAHs to cancer in humans. When PAHs from a flame mingle with nitrogen, say from a slab of meat, they can form nitrated PAHs, or NPAHs. NPAHs are even more carcinogenic than PAHs in laboratory experiments. The reasonable conclusion is that grilling meat may be hazardous to your health.

The evidence linking cancer to cooking meat over a combustion source has been accumulating for decades. Epidemiologists first noticed a connection between the consumption of smoked foods and stomach cancer in the 1960s. Japan, Russia, and Eastern Europe, where smoking is a popular way to preserve meat and fish, became laboratories for gastric cancer research. Newer studies suggest that eating smoked meats may lead to cancer even outside the gastrointestinal tract. A 2012 study, for example, linked smoked meat consumption with breast cancer.

Palmer then compares current attitudes toward grilled meats to outdated acceptance of tobacco use:

In the mid- to late-19th century, doctors determined that lip and tongue cancer rates were higher among smokers of pipes and cigars. Despite this link, major medical journals mocked those who opposed smoking. The Lancet, the leading journal of the time and still one of the most important medical journals in the world, wrote in 1879, “We have no sympathy with prejudices against … tobacco, used under proper restriction as to the time and amount of the consumption. … A cigar when the mood and the circumstances are propitious [is] not only to be tolerated, but approved.” Moderation, not abstinence, was the order of the day.


It wasn’t until 1964 that the Report of the Surgeon General finally and firmly declared that smoking was indisputably linked to the surge in lung cancer. By that time, epidemiologists had a complete picture of the long-term effects of the increase in cigarette smoking that began around the time of World War I. The conclusions practically drew themselves. Still, it took the public health community decades to agree that smoking in moderation is a terrible idea.

I learned of the article from my friend Jeff Woodhead on Twitter, who took Palmer to task for sensationalizing the dangers of grilling and noted that habitual cigarette use carries far greater risks than exposure to charred meat. I don’t disagree. However I want to defend Palmer on one point. He is actually correct to compare grilling meats to smoking tobacco, though not in the way he realizes.

That pack-a-day consumption of cigarettes greatly elevates one’s risk of lung cancer is no longer disputed by any sensible person. Moderate use of other forms of tobacco, in contrast, carries much less danger. The Lancet article that Palmer mocks for suggesting that moderation in tobacco use is nothing to worry about overstates the case but was not too far from the truth.

Just how dangerous is it to enjoy an occasional cigar “when the mood and the circumstances are propitious?” A study in Preventive Medicine compared rates of lung cancer among smokers of various kinds of tobacco to those of non-smokers. Cigarette smokers were 16 times more likely than never smokers to get lung cancer. Smokers of cigars only, pipes only, and cigars and pipes all fared much better. Further, lung cancer among pipe and cigar smokers was concentrated among those who were the heaviest consumers. “Among pipe and/or cigar smokers only, patients with lung cancer were more likely than controls to have been long-time smokers of 5 or more cigars or 5 or more pipefuls per day and to have inhaled. The odds ratio for those smoking 5 to 9 cigars or pipes per day was 3.2 and for those smoking 10 or more units 6.7. The odds ratio of those cigar or pipe smokers who inhaled was 12.3.”

A cohort study published in The New England Journal of Medicine followed about 17,000 men enrolled in Kaiser Permanente health plans who reported never smoking pipes or cigarettes. Over a course of more than 20 years, the study compared health outcomes for non-smokers and smokers of cigars. Cigar smokers carried a relative risk of cancer in the aerodigestive tract of 2.02 and in the lungs of 2.14.

A third study in BMJ examined the risks of dying from three smoking-related diseases among former cigarette smokers who had switched to smoking pipes or cigars. Their mortality risk relative to users of pipes and cigars who were never cigarette smokers was 1.51. To put that in perspective, their relative risk compared to people who had never smoked at all was 1.68. In other words, the study found that even former cigarette smokers who switch to pipes and cigars lowered their mortality risk to a level not much above that of never smokers.

Brad Rodu, an advocate of harm reduction approaches in tobacco control, summarizes many of these risks. What it basically comes do is this: Cigarettes are uniquely dangerous because they are inhaled directly into the lungs and are very addictive. Other forms of tobacco that are mainly enjoyed in the mouth and lend themselves less easily to habitual consumption are significantly less dangerous. The risks are real, but much, much lower than those associated with cigarettes.

How does all of this compare to eating meat? Palmer doesn’t cite many sources that specify the dangers, but one study he links to does associate consumption of smoked meat with breast cancer. It finds an adjusted odds ratio of 2.31-3.13. If that’s accurate (and it may be on the high side), then eating lots of grilled meat may actually be a lot like enjoying an occasional cigar — which is to say that it’s a reasonable choice many adults may decide to make.

(Note: Some of the studies cited in this post use relative risk while others use odds ratios. These are not identical measures but should be roughly comparable.)

It’s also interesting to ask how this compares to the dangers of exposure to secondhand tobacco smoke. Smoking bans have proliferated on the justification that workers should not be put at risk. Whatever one’s opinion of these policies, consider the findings from the landmark 2006 report from the Surgeon General. The report concludes without equivocation that environmental tobacco smoke increases the risk of lung cancer in non-smokers. By how much? The reports’ table of meta-analyses of studies estimates relative risks of exposure at home for non-smoking spouses or at work for non-smoking employees in a range of 1.12 at the low end to 1.43 on the high end. Those are low relative risks! If they are enough to justify comprehensive bans on indoor smoking — not to mention the outdooor bans that are now so popular — then one may well question the ethics of allowing restaurant cooks to expose themselves to grill smoke on our behalf.

One of the big problems with reporting on cancer is that it focuses on the wrong question. Journalists ask, “Does X increase the risk of cancer?” The answer is very often yes, but they don’t follow-up with, “By how much?” Lifestyle choices carry trade-offs and better reporting would help readers put them into perspective. Palmer, to his credit, does write that the “risk-reward equation for smoking differs from that of grilling or frying meat” and acknowledges that the epidemiology on the latter is not yet clearly established. His mistake is to carelessly lump all forms of tobacco use into one, ignoring the fact that different types and different use patterns carry substantially different levels of risk.

Of course, few yet advocate completely giving up grilled or smoked meat, much less passing legislation to restrict it. “Everything in moderation,” said one of the toxicologists quoted in the Slate article. One assumes that the toxicologist’s definition of “everything” doesn’t include tobacco, because who would say that these days? But Palmer’s comparison of grilling to smoking isn’t crazy. It just tells us a lot more about contemporary bias against tobacco use than it does about the dangers of meat.