The Fall of Roman Heart Attacks

An article in the American Heart Association journal Circulation reports that acute coronary events declined in Rome during the first year after Italy's ban on smoking in public places took effect. Looking at hospital discharge records and causes of death, the researchers found that the age-adjusted rate of acute coronary events among 35-to-64-year-olds was 11 percent lower in 2005 than the average for the five years before the ban. The reduction among 65-to-74-year-olds was 8 percent. The researchers attribute these decreases to the smoking ban. "Smoking bans in all public and workplaces result in an important reduction of acute coronary events," says one of the authors. "The smoking ban in Italy is working and having a real protective effect on population health."

A reduction of 8 percent to 11 percent in heart attacks is nothing to sneeze at, but it's a far cry from the 40 percent reduction anti-smoking activists claim occurred in Helena, Montana, as a result of the smoking ban there. (Trying to explain the inconsistency, the authors of the Italian study emphasize the Helena study's weakness, noting that it "counted only 24 hospital admissions for AMI [acute myocardial infarction] from a small community in Montana in the 6 months after the ban.") Attributing even the relatively modest changes seen in Rome to the smoking ban is iffy, since heart attacks were already declining there before the ban. 

In the younger age group, the acute coronary event rate went down from 2000 to 2001, up from 2001 to 2002, and declined steadily thereafter. In the older group, the rate rose steadily from 2000 to 2003, then declined for two years in a row. As Michael Siegel notes on his tobacco policy blog, in both cases there were single-year reductions prior to the smoking ban at least as big as the drop the year afterward. The older group saw a 6 percent drop between 2003 and 2004 (before the ban), the same as the drop between 2004 and 2005 (after the ban). The younger group saw an 8.5 percent drop between 2002 and 2003, compared to a 6.3 percent drop between 2004 and 2005.

So even on their face, these data do not indicate that the smoking ban was associated with an extraordinarily large reduction in heart attacks. If they did, there would still be the question of whether the smoking ban caused the reduction or just happened to precede it. As the researchers concede, the absence of data from a comparable jurisdiction without a smoking ban is a major weakness in the study. Whether as a result of pre-existing (or coinciding) trends or simply by chance, heart attacks are bound to decline in some places after smoking bans take effect; the question is whether they are more likely to fall (or tend to fall more rapidly) in such places than in otherwise similar locations without smoking bans.

I have no doubt that smoking bans, over the long term, will tend to reduce heart attack rates to the extent that they discourage smoking. (The smoking rate in Rome fell from 35 percent to 31 percent among men the year after the ban while remaining about the same among women.) I just think it's highly implausible that the effect would be seen immediately, since epidemiological studies of former smokers show it takes years for their heart disease rates to decline. It's even more implausible that the effect would be due primarily to reductions in exposure to secondhand smoke, as the authors of this study argue. They cite the commonly quoted estimate that secondhand smoke raises the risk of heart disease by 30 percent. But that estimate is derived from epidemiological studies of long-term, intense exposure, mainly among people who live with smokers for decades. Assuming the association indicates a cause-and-effect relationship, you would not expect to see a noticeable decline in risk after one year of smoke-free bars and restaurants.

Even if underlying disease rates don't change that quickly, maybe heart attacks fall because the prevalence of a potential heart attack trigger for people who already have heart disease is reduced. Taken at face value, the numbers from this study imply that something like one in 10 heart attacks among 65-to-74-year-old Italians prior to the ban occurred when they keeled over in smoky bars or restaurants as a result of inhaling tobacco combustion products. I suppose that might be true. But where's the evidence? It's not in this study.

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  • Steve Verdon||

    Looking at hospital discharge records and causes of death, the researchers found that the age-adjusted rate of acute coronary events among 35-to-64-year-olds was 11 percent lower in 2005 than the average for the five years before the ban. The reduction among 65-to-74-year-olds was 8 percent.

    Let me see, they are comparing a single year to a five year average. If I look at the average of 5 six sided rolls of a die and then compare it to one year....

    I guess my first question is how much variability is there from year-to-year. I'm thinking this data doesn't show what the authors are claiming.

  • Paul||

    Chalk one more up for the International War on Drugs.

  • Paul||

    If I look at the average of 5 six sided rolls of a die and then compare it to one year....

    Your Dexterity and Charisma would be teh pimp...

  • ||

    Shouldn't the headline be something more like, "the decline and fall of Roman heart attacks"?

  • ||

    1 unit one year
    3 units the second year
    Average 2 units

    Doesn't take a rocket scientist to figure that one out

  • ||

    Jesus - the denialism here is palpable. Opposing a smoking ban on the perfectly respectable grounds that we're talking about grown-ups is one thing, but comparing it to activists in a small Montana town is another.

    Embarrassing, really. No one denies that smoking is bad for you and can cause lung cancer and heart disease. Posts like this imply that your objection to smoking bans is based on "sound science" or something like that, which is the losing argument since it forces you to trash perfectly reasonable analysis like this article.

    Stick to the winning argument about choice and save those amateurish stat skills for the real hack studies.

  • green eye||

    Bravo, Matt. Anyone who has been a smoker and a non-smoker knows the difference with regard to heart and lungs.

  • robc||

    Matt,

    There is a difference between opposing a ban based on "sound science" and opposing the people who favor a ban because they are using "bogus science".

    I can oppose a ban based on choice and also rip apart the "science" of the people using a ban if they are trying to push it based on science. The fact is that 1 year of data is bogus. Come back with a decade of data and we can talk. Yes, that means delaying future bans based on science for 10 years, but so what? You dont get to use science as a reason until you have good science as a reason.

  • ||

    Jesus - the denialism rationality here is palpable.

    Yeah, how we dare we bring up shit like, you know, facts and logic to challenge the fashionable prejudices of the day.

    No one denies that smoking is bad for you and can cause lung cancer and heart disease.

    You can easily debunk junk science without also rejecting good science. In fact, accepting junk uncritically is what is truly non-scientific.

  • ||

    Matt -- Ya think maybe possible it's to do both?

    Tho' at this point any type of resisitance is just spitting into the wind of authoritarian hegemony.

  • ||

    I wonder if cigarettes were banned completely, how many people would still smoke? Is it 50% of current levels? More? Less? We know one thing, the price would skyrocket.

  • ||

    I wonder if cigarettes were banned completely, how many people would still smoke?

    There would be some kind of market, maybe black, in concentrated alternatives. I don't know what the crack version of cigarettes would be, but we'd sure find out.

  • Steve Verdon||

    Your Dexterity and Charisma would be teh pimp...

    Well I was going to use a d20, but then i thought that would be too obivous.

    Matt,

    I'm not sure what point you are trying to make. Nobody, me, Sullum nor anybody else that I know, is claiming smoking is not bad. But this article that Sullum is pointing to is not sound science. Comparing one years observation, to the average over several years is not invalid, but in this case we have no way of knowing if there is reason not to chalk this up to "confounding factors". It sure as Hell isn't evidence in favor or smoking bans. That evidence will likely come in several more years.

    And when it comes to sound science it isn't about winning arguments, but pushing sound science. You have it backwards, I'm afraid.

  • Michael J. McFadden||

    Matt wrote, "Opposing a smoking ban on the perfectly respectable grounds that we're talking about grown-ups is one thing, but comparing it to activists in a small Montana town is another."

    Matt, the lies based upon the "activists in a small Montana town" were used to fuel smoking bans all over the United States throughout the past five years. See:

    http://www.acsh.org/factsfears/newsID.990/news_detail.asp


    If an 11% drop today proves that a smoking ban reduces heart attacks, wouldn' the 8.5% drop between 2002 and 2003 prove that NOT having a smoking ban reduces heart attacks?

    Or does "proof" only become "proof" when an Antismoker says it is?

    Michael J. McFadden
    Author of "Dissecting Antismokers' Brains"
    www.TheTruthIsALie.com

  • ||

    Sorry I tuned out, but on the off chance anyone checks back into this post:

    First, I'm obviously disagreeing with the idea that this is a hack study. A) The authors average previous years' heart attack incidence to lessen the effect of random fluctuations. B) The authors control for plausible contributing effects that they can think of, such as air pollution. C) The city of Rome is relatively large with many heart attacks each year. D) The paper passed peer review.

    Reasons why it might not be completely accurate include other confounders that they fail to account for and the presence of only one year of post-ban data. However, this is their best point estimate.

    One of your objections seems off base to me in particular (and is the basis for my claim that you don't understand statistics). You maintain that a fall in incidence in pre-ban years weakens their findings, when in fact the very contributing effects they account for could be responsible (and wouldn't bias their result).

    This isn't a hack study, it is a hack criticism. Unintentionally so, probably, but Reason bloggers have thick skin so I won't mince words. But I have nothing backwards here. We already know what the end result of these studies is going to be, we just aren't sure of the magnitude. So don't pin your criticism of smoking bans on uneducated critiques of peer-reviewed work by trained statisticians. (Get libertarian statisticians to make those critiques!)

    This does look like a case of futile spitting into the wind, but that's because Mr. Sullum isn't a very good spitter and he isn't using his uzi (which was the argument I suggested).

    A 5-10 percent reduction in heart attacks isn't necessarily worth infringing on peoples' rights, even to non-libertarians, after all ...

  • ||

    It's even more implausible that the effect would be due primarily to reductions in exposure to secondhand smoke, as the authors of this study argue.

    Second hand smoke? In Rome?! I thought everybody in Europe smoked.

GET REASON MAGAZINE

Get Reason's print or digital edition before it’s posted online

  • Video Game Nation: How gaming is making America freer – and more fun.
  • Matt Welch: How the left turned against free speech.
  • Nothing Left to Cut? Congress can’t live within their means.
  • And much more.

SUBSCRIBE

advertisement