Scottish Smoking Ban Study Leads to Huge Drop in Journalistic Credibility

Jacob Sullum | November 21, 2007

Two months ago, a study reporting that the imposition of Scotland's smoking ban had been followed by a 17 percent drop in hospital admissions for heart attacks prompted credulous headlines like these:

Smoking ban brings big cut in heart attacks in Scotland, study finds (The Guardian)

Smoking ban 'reduces heart risk' (BBC News)

Scottish smoking ban cuts heart attacks (The Telegraph)

Scottish Smoking Ban Leads to Huge Drop in Heart Attacks (Der Spiegel)

Now it turns out that the complete admissions data for the year following the ban show a drop less than half as big as the one claimed in the study and in all those news stories. And as Michael Blastland shows in a recent BBC News article, the 8 percent drop may simply be part of a long-term trend:

Heart attacks have been falling steadily for some years now. The percentage falls in the three years before the ban were 5.1%, 4.7% and 5.7%. So the fall since is still bigger than the trend would lead us to expect, but bigger only by about three or four percentage points—an improvement, but retreating fast from the magnitude of 17.

The latest release also makes clear that even an 8% fall in heart attacks is not unprecedented. There was another, larger drop between 1999 and 2000 of about 11%.

This seems to demonstrate significant variability around the trend, suggesting that last year's 8% drop might even be the result of chance. It is conceivable, although perhaps unlikely, that the smoking ban had no effect at all. The figures could be a result of no more than the ordinary ups and downs of statistical variation from one year to the next.

Reporters can't be expected to look at data that haven't been released yet. But the idea that a smoking ban could cause an immediate, dramatic drop in heart attacks, whether by reducing secondhand smoke exposure or by encouraging smokers to quit, is so scientifically implausible that journalists should be automatically skeptical. Random variation and/or continuation of pre-existing trends are much more likely explanations for a decrease in heart attack admissions that happens to follow the imposition of a smoking ban. In the absence of more evidence, including additional years of followup and data from other jurisdictions with smoking bans (most of which apparently have not seen the big changes found in the few places that anti-smoking activists like to cite), the breathless reports prompted by studies like this one are absurdly premature.

[via The Rest of the Story]

Comments

|11.21.07 @ 12:50PM|#

But the idea that a smoking ban could cause an immediate, dramatic drop in heart attacks, whether by reducing secondhand smoke exposure or by encouraging smokers to quit, is so scientifically implausible that journalists should be automatically skeptical.

No, it's not. It's entirely plausible to think that people who already have heart problems could have more heart attacks when exposed to second-hand smoke.

It only looks implausible if you ignore the difference between chronic effects and acute effects.

|11.21.07 @ 12:52PM|#

17% is pretty large, though.

|11.21.07 @ 12:52PM|#

How silly! They clearly don't have any clue that what they're positing just might be a false correlation.

Besides, everyone knows that it's the sword ban, not the smoking one, that has reduced the incidence of heart attack. Let me tell you, seeing a claymore in your gut is a surefire way to stopping your heart.

|11.21.07 @ 1:03PM|#

I attribute the drop to fewer German nihilists hanging outside the bowling alleys.

|11.21.07 @ 1:07PM|#

Shut the fuck up, Jamie. You're out of your element.

mark|11.21.07 @ 1:08PM|#

Anybody who reads the newsfeeds reguarly can see that we have replaced investigative journalism with cut n paste journalism and i presume when information comes from certain sources it is accpted as fact.
A modecum of internet research on SHS shows some very interesting and widely differing oppinions,especially from bodies not funded by tobacco or nicotein replacement products.

ed|11.21.07 @ 1:09PM|#

It's entirely plausible to think that people who already have heart problems could have more heart attacks when exposed to second-hand smoke

Yes, it is. If you're a gullible serf. But if by "exposed" you mean eight hours a day in a closed car with a cigar smoker who's doing six sticks a day--then yeah, there might be a slightly increased risk of triggering a cardiac episode. And vomiting. And a headache. And the sniffles. Or nothing at all.

Neu Mejican|11.21.07 @ 1:13PM|#

Anyone else find it ironic when Jacob Sullum posts on the topic of "Journalistic Credibility?"

(c.f., http://www.reason.com/blog/show/123575.html)

|11.21.07 @ 1:25PM|#

Anyone else find it ironic when Jacob Sullum posts on the topic of "Journalistic Credibility?"

Nope. Anyone find it ironic when Neu Mejican posts snarky comments under the guise of having anything of substance to say?

|11.21.07 @ 1:29PM|#

I have a rule:
If it's in the news, it ain't science. This rule has never failed me when it comes to any sort of scare, or whatever. "Eleventy billion people die from obesity every year." "Drink more caffeine, die early." "Drink more caffeine, live longer". etc.

Rhywun|11.21.07 @ 1:40PM|#

The local channels here in NYC have replaced the "don't smoke or you'll get a hole carved in your throat like me!!" spot with a new one, "don't smoke or you'll get diabetes, two heart attacks, and assorted other ailments, necessitating various amputations and whatnot, before the age of 30!!!". I wonder at what point does the increased hysteria generate diminishing returns? Ex-smoker Bloomberg at al. are already complaining that smoking isn't declining as fast as they "want" it to. How can that be with such "effective" commercials...?

|11.21.07 @ 1:41PM|#

andyinsdca, you fool!
That would put you in a state of perpetual skepticism, when all you need is more government grants for further studies!
You fool!

Guy Montag|11.21.07 @ 1:42PM|#

So the Scotts have found the fountain of youth without the assistance of a science-fictional engine operator?

No, wait, this is almost the same thing. The fix to all of the world's ills is in the keyboard of someone with a writing degree.

|11.21.07 @ 1:43PM|#

Scottish Smoking Ban Study Leads to Huge Drop in Journalistic Credibility

Credibility can really be less than zero? This is only in theory, right? Certainly some nonnegative number must at least be within the error bars of the empirical credibility measurement, no?

Or perhaps journalistic credibility only appears to be zero, but measuring it on a logarithmic scale shows that it really can drop further.

highnumber|11.21.07 @ 1:46PM|#

Neu Mejican,

Too oblique. No idea what the point is. Left unable to use personal pronouns.

|11.21.07 @ 1:51PM|#

I'm no scientist or anything, but I thought smoking takes a long time to have adverse effects. Or at least, longer than a year anyway. Goebels would be proud...

|11.21.07 @ 1:55PM|#

Taktix®,

So, what you're saying is that the Nazis should've have simply smoked a lot around Jews, without all the overt antisemitism, prejudicial laws, and concentration camps. You're an insidious thinker.

Paul|11.21.07 @ 1:57PM|#

And the drug war continues...

Ok, kids, what else can we ban to improve public health?

Oh, and joe, yes, it is implausible. Highly implausible. Reducing exposure to 2nd hand smoke in one venue will not produce a 17% (or anything close to it) drop in heart attacks.

|11.21.07 @ 2:01PM|#

Ok, kids, what else can we ban to improve public health?

I bet banning haggis would be the best public health ban the Scots could undertake.

;P

(BTW: I'm a MacDonald on my mother's side.)

|11.21.07 @ 2:08PM|#

Aresen! Haggis isn't just good for the body, it's good for the soul.

Here's a haggis recipe. If you can't cook, buy some canned haggis.

Of course, deep-fried haggis might have some negative effect on health if consumed in mass quantities.

|11.21.07 @ 2:10PM|#

Aresen,

We cosmopolitan Urkoboldians posted something on Canadian Thanksgiving today. Explained by a New Zealander, natch.

Neu Mejican|11.21.07 @ 2:15PM|#

Anyone find it ironic when Neu Mejican posts snarky comments under the guise of having anything of substance to say?

How is snark not substance?

JS, basically a propagandist, is discussing the journalistic credibility of other journalists. I find that ironic.

I guess I should stick to comments about "German nihilists hanging outside the bowling alleys."

Neu Mejican|11.21.07 @ 2:16PM|#

Highnumber,

Personal pronouns are overrated.

Check out Sullum's comment on the DSM in regards to the article he is commenting on.

Paul|11.21.07 @ 2:19PM|#

Taktix:

You don't have to be a scientist to understand statistics, margins of error, statistical anomalies, dampening of said anomalies over time, or contextual health dangers vs. immediate health dangers.

For instance, if we banned the ingestion of strychnine and handling of rattlesnakes, we might see an immediate drop in related health effects. But to ban smoking or transfats in a public venue and then suggest that within one year we immediately saw a 17% reduction in heart attacks (a health effect related to hundreds or even thousands of environmental and genetic factors) is near plain stupidity.

First, it would suggest that smoking is pretty much the only damned cause of heart disease. Second, assuming the first, it would assume that if you just banned cigarettes completely, in both public and private venues, you know, like we do cocaine, marijuana, heroin etc., then you'd see a mind-boggling drop in heart attacks. Oh, sorry, an even MORE mind-boggling drop in heart attacks. I would estimate you'd triple that percentage drop.

When my Dad was around 80 years old, he had quintuple bypass. I pressed one of the surgeons for information as to how and what my dad could to do avoid any future problems after his surgery-- in other words: what caused this, and how can he prevent it?

The surgeon, widely considered one of the better in his field responded: There are many correlations to heart disease, but the fact of the matter is, despite what you read in the press, we simply don't know what causes it.

Now admittedly this was the mid-nineties. Maybe there's some research somewhere that says we cracked this nut, and 90% of all heart attacks are smoking related?

And if you actually read the article (a painful endeavor I must warn) they report... I hesitate to have these pixels rasterized on my screen, but here goes...

Non-smokers benefit most with 20% fall in first year

20%. Non-smokers. Marinate on that for a bit. Let that soak in. Sitting near to someone for a couple of hours a week that lights up drops your heart attack risk by 20%. Wow, imagine the drop if we could just ban cigarettes altogether? Sure even if true (can you tell I'm laughing) we would see statistical diminishing returns as we expanded the bans. But when you think about it, banning smokes in pubs and restaraunts is only a first relatively small step. There are plenty more venues which play larger roles in our lives than pubs and restaraunts, so those effects of diminishing returns won't be seen for a while.

In summary, if we banned cigarettes completely, going as far as making posession a crime, what, pray tell would we actually die from?

R C Dean|11.21.07 @ 2:20PM|#

It's entirely plausible to think that people who already have heart problems could have more heart attacks when exposed to second-hand smoke.

If exposure to "social" second hand smoke over years doesn't cause heart problems (and I don't think anyone has shown that it does), then now on earth could one-time exposure to second hand smoke cause a heart attack?

|11.21.07 @ 2:24PM|#

R C,

Due to the nonsmoker's indignation over having to endure the smoker's presence?

|11.21.07 @ 2:50PM|#

No, it's not. It's entirely plausible to think that people who already have heart problems could have more heart attacks when exposed to second-hand smoke.

Yes, I believe it is entirely plausible that people would *think* so. Why, it DID happen: journalists thought the drop could be attributable to the ban. However, the problem is that the dramatic drop in itself sounded too good to be true. It was the dramatic drop itself that sounded implausible.

Dave T.|11.21.07 @ 3:10PM|#

Dear Journalists,

Please collect some micro panel data, run a DND regression controlling for various factors (age, sex, race, etc.) with the Scottish as your treatment, and some similar region w/o a smoking ban as a control, and let us know when you have some real results with significant coefficients.

Thanks.

Paul|11.21.07 @ 3:13PM|#

Not to pile on, but...

A cancer researcher who is in favor of "vigorous smoking bans" has this to say about these dramatic effects on health-- in particular the Helena, MT ban which resulted in a 40% drop in heart attack admissions:

Finally, the "immediate effect" should make anyone stop and question the connection the authors are asserting. There are few interventions in public health that have such an immediate impact. Even if all active smokers in Helena had quit smoking for at least a year, one would not expect to see such a dramatic effect.

Read it here.

|11.21.07 @ 3:45PM|#

Pro Libertate | November 21, 2007, 2:10pm | #

Aresen,

We cosmopolitan Urkoboldians posted something on Canadian Thanksgiving today. Explained by a New Zealander, natch.

I'm at work, so I couldn't view the video.

The real reason Canadians celebrate Thanksgiving in October is that it's too bloody cold here up here to play football in November. [Except for our finest second-rate American Mercenaries players*, who battle for the Grey Cup this coming weekend while we Canadians sit in the stands (or our living rooms) and drink rye whiskey to stay warm.]

*Who, immediately after the game, head home to Georgia to spend the next 6 months thawing out and recuperating from frostbite.

Happy Turkey Day, BTW. I'll be working while you're watching the games.



Back to haggis: I firmly believe that haggis was invented by the Scots to poison the Sassenachs.

Paul|11.21.07 @ 4:33PM|#

Aresen, I can't resist... are your second-rate American players first-rate in Canada?

He who can't make it in America, totally dominates in Canada?

|11.21.07 @ 4:51PM|#

I've learned a lot from this thread.

For example, there are a lot of people who don't understand the difference between acute and chronic effects.

And they feel quite confident opining about medical matters.

Every single one of the retorts I've gotten looks like it's come from people who've never so much as taken high school biology, and I'm not responding to any of you. Global-warming-thread quality stuff. You should be proud.

|11.21.07 @ 4:53PM|#

Paul,

The annual rate of non-smokers having heart attacks is incredibly small. A 20% drop, when we're talking about a chance that starts out at, say, 0.020% (one change in five thousand) reduces that figure to a 0.016% chance. That doesn't strike me as unbelievable.

|11.21.07 @ 4:56PM|#

Paul

Well, considering that CFL salaries run about 10% of NFL salaries, how many Hiesmann (sp?) Trophy winners do you think there are in the CFL?

Well, gee. That offer of $8 million a year to play in Dallas just couldn't compete with $750,000 a year to freeze my ass off in Regina.

not

Franklin Harris|11.21.07 @ 5:16PM|#

Aresen! Haggis isn't just good for the body, it's good for the soul.

I've tried haggis only once, from a booth that a Scottish restaurant set up at a Nashville "cultures of the world" festival. Is it supposed to taste like greasy meatloaf without the tomato sauce?

highnumber|11.21.07 @ 5:31PM|#

greasy meatloaf without the tomato sauce

That sounds good!

|11.21.07 @ 5:35PM|#

Is it supposed to taste like greasy meatloaf without the tomato sauce?

I see you got top quality haggis.

Most of it tastes like plaster in a baggie, cooked for several hours and basted with used engine oil.

Paul|11.21.07 @ 6:23PM|#

Every single one of the retorts I've gotten looks like it's come from people who've never so much as taken high school biology, and I'm not responding to any of you. Global-warming-thread quality stuff. You should be proud.

So then you disagree with the cancer researcher who has presumably taken the odd course in biology, then?

The annual rate of non-smokers having heart attacks is incredibly small. A 20% drop, when we're talking about a chance that starts out at, say, 0.020% (one change in five thousand) reduces that figure to a 0.016% chance.

Joe, I know. I not only know this, but I understand it as well. I don't deny that there was a 20% drop in heart attack rates. I have to assume that the journalist didn't lie about the figures. It's the 20% rate in heart attacks directly and unquestioningly attributed to a year-long smoking ban in "enclosed public spaces". That, and that alone is the point of this debate. Its basis is wanting.

I'm well aware of your suggestion that a non-smoker with the chronic condition of heart disease may suffer an acute heart attack based on the limited exposure to second hand smoke. But that's not apparent in these stories or assertions. They're merely applying post hoc ergo propter hoc reasoning.

Paul|11.21.07 @ 6:24PM|#

freeze my ass off in Regina.

Watch it, Aresen, this is a family joint.

|11.21.07 @ 6:31PM|#

Paul,

You might have noticed, I responded to YOU immediately after that comment. You raise some good points.

All I'm saying here is, a 20% drop in heart attack rates among nonsmokers is, in and of itself, not such a dramatic change that it makes the claim absurd.

Paul|11.21.07 @ 6:38PM|#

Look at it this way, joe.

Smoking rates have been steadily dropping. One could further presume that with the rate of smokers dropping steadily, the number of people exposed to second hand smoke is dropping exponentially.* Have heart attack rates amongst non-smokers dropped in direct correlation with this? That might be something to ponder.

*I say exponentially because it is presumed that one smoker exposes multiples of non-smokers to second-hand smoke. So for every smoker that stops, that's many smokers who receive less or no second-hand smoke (depending on their relationship to the smoker). All of this is of course arguable.

You might have noticed, I responded to YOU immediately after that comment. You raise some good points.

And I appreciate that, joe. I may not always know what I'm talking about(?!), but I try to be honest in my posts.

Happy holidays/winter festival/harvest festival to you and yours. Seriously.

highnumber|11.21.07 @ 7:13PM|#

plaster in a baggie, cooked for several hours and basted with used engine oil

That sounds bad!

Robert Feal-Martinez|11.22.07 @ 6:44AM|#

The reality is smoking bans are based on a provable lie. Passive Smoking is just an irritant. The public and politicians have been gullible enough to believe the Big Drug company funded junk science, this is just another of these, which has been shown to so. The detail of the research will never see the light of day.

|11.22.07 @ 7:48AM|#

I wrote to Professor Pell on the day this garbage came out. I asked for a copy of the report and posed to her questions about peer review, why January and February were missed out, what role Statins have in reducing heart attacks (25%)etc, etc. Alas not even the courtesy of an aknowledgement. If you are better connected try: [e-mail]

Neu Mejican|11.22.07 @ 11:37AM|#

the Big Drug company funded junk science

So the big drug companies have more clout than big tobacco now?

A slug fest between giants!

8^)

|11.23.07 @ 8:18PM|#

xiexie
好秘书 呼吸网 肿瘤网