Smoking ban reduces heart attacks…more bollocks

My instinctive reaction to this story was that it\’s bollocks:

The first study to examine the possible effects of the ban on health in England suggests that there were 1,200 fewer heart attack admissions — a 2.4 per cent drop — in the year after it was imposed. The reduction in emergency admissions saved the NHS an estimated £8.4 million and may have prevented almost 200 deaths.

Public health experts said that the findings, published in the British Medical Journal, clearly demonstrated the effectiveness of anti-smoking legislation.

And whadda ya know? It is bollocks.

As you can see, the decline in admissions in the year after the smoking ban was larger than the year before but smaller than the year before that. In fact, the average in the previous two years was 4.2%—almost exactly what it was in the year after the ban (4.26%).

Admissions for heart attacks have been falling at roughly the same rate for years. There is no blip, increase or improvement in the year after the ban.

We are, I\’m afraid, being lied to.

Again.

31 comments on “Smoking ban reduces heart attacks…more bollocks

  1. The BBC just do as their told.
    I don’t think a real journalist exists within that Quango.
    They should be ashamed spinning lies that .
    Oh hang on they spin the Global Warming lie as well do they not.
    Problem is the dummies believe it.
    At the moment the ratio of dummies to free thinking people is roughly 50/50.

  2. James Le Fanu has argued for years that the rise and subsequent fall of heart disease looks like the classic behaviour of an infectious disease.

  3. On the Beeb-bashing: the BBC’s job, as with any news organisation, is to report that the study in question has been released, and to highlight serious criticisms of it. It is not, in the absence of glaring and obvious flaws, to peer-review the study.

    There are no glaring and obvious flaws in the latest study (Snowdon’s criticism of it is based on “the survey assumes that the fall in heart attacks was slowing, which I don’t think is likely because I just don’t, and therefore it must be rigged”. This is not an argument so much as a demonstration that the person making the criticism is an idiot), so the correct approach is for the BBC to report it.

    If someone newsworthy were to criticise the study (which doesn’t include a random blogger hurling faeces), or if subsequent published research were to show it was in some way flawed, then the correct approach would be for the BBC to report the criticism, and to try and ensure that the final report reflected the balance of informed opinion.

    The earlier studies that suggested 20%+ declines *were* revealed to be bollocks (and really obvious, stupid bollocks – the credibility of the researchers responsible for those ones, which is not the same as the team here – has to be in question). That’s why the commentary piece that Belinda highlights exists.

  4. johnb: this study is clearly a case for stopping the large scale prescribing of statins, and any other interventions as the fall has been attributed solely to the smoking ban.

  5. JohnB… Maybe it is the BBC’s job to report the study.

    However, if it’s to become a headline item a simple soul such as me would expect that the study should undergo just a tad of scrutiny by a “science” correspondent (yes I know that the BBC’s science “experts” all have degrees in English, journalism or medieval history, but they could at least ask someone better qualified).

    That it’s impossible to distinguish from either the graph or the table of steadily-falling “heart attack” admissions exactly where the smoking ban came into effect would suggest that it’s made toss-all difference.

  6. What fall?

    http://www.isdscotland.org/isd/5777.html

    this page has the routine stats listed in tables for Scotland (MC and AC) (where the smoking ban was introduced in 2006). I can find no drop in heart attacks here that is greater than drops occurring before the smoking ban took place. Drops have been occurring for years for many reasons.

  7. Ian: no, out of the 4.3% fall, 2.4% has been attributed to the smoking ban. The rest is made up of the assorted other ongoing demographic and public health efforts.

    Pogo: if you have a graph on a multi-year scale like that one, then sizeable year-on-year growth variances don’t show up. You’d actually need the kind of, wildly implausible, 20% decline the earlier studies suggested to make a dent in the chart.

    Belinda: heart attacks continued their existing falling trend after the bans, in both Scotland and England. The analysis linked here concluded that in England, the ban was partly responsible for continuing that trend despite pressures that might have slowed it.

    If this study had said “heart attacks fell, therefore the smoking ban helped”, it would have been stupid. If it had said “heart attacks continued to fall, therefore the smoking ban didn’t help”, it would have been stupid. The fact that it tried to split out the different components behind the fall shows that – while it may in future be shown to have been wrong – it wasn’t stupid.

  8. JohnB – What it should have said is “heart attacks continued to fall at much the same rate as before, therefore it’s impossible to quantify any contribution made by the smoking ban”.

    Ok, it’s a multi-year graph and a statistically-insignificant change isn’t going to be visible, but in light of the figures used for the report:-

    2002/03: 61,498
    2003/04: 60,680 (a fall of 1.33%)
    2004/05: 58,803 (a fall of 3.1%)
    2005/06: 55,752 (a fall of 5.19%)
    2006/07: 53,964 (a fall of 3.21%)
    2007/08: 51,664 (a fall of 4.26%)

    … the 4.26% total reduction and 2.4% “smoking-ban-related” reduction are not outside the bounds of the natural variation of previous years.

    The fact that it tried to split out the different components behind the fall shows that – while it may in future be shown to have been wrong – it wasn’t stupid.

    I’m afraid I disagree – it might not have been “stupid” per se, but it’s “voodoo science” at best and deliberate dissimulation at worst.

  9. As far as I can see the smoking ban changed only where people were smoking. It has not been shown that fewer people smoked. Jill Pell’s figure for Scotland showed that the drop in heart attacks for smokers fell by 14 per cent, again a truly stupendous drop that did not find its way into the normal stats. A report in September showed that the smoking rate in Scotland barely dropped at all, which means that the smoking ban made virtually no difference to smokers except where they were allowed to smoke: http://www.dailyrecord.co.uk/news/health-news/2010/02/07/exclusive-smoker-scots-refuse-to-call-it-quits-despite-ban-86908-22024861/

  10. @Belinda – careful with the voodoo studies. A flat *number of smokers* doesn’t mean a *fall in smoking*. IM(extensive)X of pubs pre- and post-ban, the number of cigarettes smoked per patron has declined significantly since having to go outdoors, even if they haven’t otherwise changed their smoking behaviour.

  11. JohnB – Re your extensive experience of pubs, echoed by my own…

    I wonder if any fall in coronary problems in smokers has been countered by an increase in pneumonias from having to brave our generally dreadful weather! 🙂

  12. How is there certainty about the amount of tobacco smoked when there is considerable illicit traffic in tobacco? I know many people who reported a rise in their own personal smoking.

  13. John B,

    On the Beeb-bashing: the BBC’s job, as with any news organisation, is to report that the study in question has been released, and to highlight serious criticisms of it. It is not, in the absence of glaring and obvious flaws, to peer-review the study.

    The problem is that you’re then no longer an unbiased news organisation. You’re giving a platform to groups that you favour, that you assume are right, or whose values you wish to promote. Which is fine when you’re talking about the Telegraph or The Times, but the BBC is supposedly unbiased.

  14. @johnb: “that’ll be because the epidemiology of the two biggest non-genetic risk factors – smoking and obesity ..”: surely that’s a hopeless explanation? Heart attacks went roaring up decades before obesity did and have been coming down while obesity is rising. So, even setting aside your attempt to imply that “risk factor” means “cause”, your argument still fails.

  15. Whatever the real truth about smoking related heart attack admissions. the more facts I read the more obvious it is that the BBC article, is at best badly written, and at worst deliberately misleading. Consider the opening two paragraphs:-

    “There were 1,200 fewer hospital admissions for heart attacks in England in the year after July 2007 – when the smoking ban came in, a study suggests.

    The 2.4% drop was much more modest than that reported in some areas where similar bans have been introduced, but may still have saved the NHS over £8m.”

    It definitely implies that 2.4% was the whole reduction, and due entirely to the smoking ban. It makes no mention there was an overall 4.3% reduction, and in fact you only learn this by going to other articles.

    “But the researchers cannot say categorically that the drop was down to a reduction in exposure to second-hand smoke, or because fewer people are smoking in the first place. Data on the smoking status of the patients admitted was not available, and the Bath University team intends to look at this next. ”

    So they admit it’s a load of old dingo’s kidneys, but hey that won’t stop them publishing, and the press reporting as fact what amounts to speculation.

    I could go on, but then I saw this article in the Daily Mail.

    http://www.dailymail.co.uk/health/article-1285201/Secondhand-smoke-increases-risk-mental-health-problems.html?ITO=1490

    The article ends with

    “‘Taken together, therefore, our data are consistent with other emerging evidence to suggest a causal role of nicotine exposure in mental health,’ the investigators concluded in their study that appeared in the Archives of General Psychiatry.”

    Plenty of foods contain nicotine. See http://www.craigsams.com/pages/tobac.html

    If I still smoked I would definitely be going outside for a fag right now.

  16. Joseph: if someone who’s a respected scientist at a respected organisation publishes a report, then an impartial news source ought to report it. If they deliberately ignore reports by equally respected organisations because one suits their political bias and the other doesn’t, then they’re not an impartial news source. The Times and DM certainly do the latter, which is completely fair enough as you say. There’s no evidence that the Beeb does, and if they did they’d be failing in their remit. If serious people in the relevant field respect the wrong organisations (which is basically the anti-AGW-coverage argument), then that really isn’t the fault of the news sources reporting on the issue.

    Dearieme: now plot smoking *and* obesity. The theory is, decreased physical activity (which is primarily what obesity is about – most people in the US/UK have always-post-industrialisation eaten crap food) is partly offsetting the fall in heart disease caused by the decline in smoking, now that most of the people who’d ever give up smoking have done so. Similarly, the public health justification for the smoking ban – which, incidentally, I don’t view as a legitimate justification on JS Mill grounds – is that it’ll make the remaining smokers smoke a bit less.

    Ian: yes, the BBC report isn’t much cop (I’d not even noticed, because I default to reading any stats-based news by checking the source). Nearly all media coverage of statistics isn’t much cop, for obvious <a href="http://www.badscience.netpaging Ben Goldacre reasons. However, the way in which they fuck up statistics is seldom related to their political bias – I’ve had data-driven press releases I wrote published in the lefty press by journalists who’d misunderstood the data in ways that made their articles slant to the opposite view from their paper’s line, and published in the right-wing press by journalists who’d misunderstood the data in ways that did exactly the same.

    (after some fiery baptisms, I’m now *very good indeed* at writing press releases featuring data, and at understanding how journalists will get any available ambiguities wrong. Because I’ve always worked for impartial research companies who’re reliant on credibility to maintain their reputation, this power has been largely used for good not evil. The press release here definitely strikes me as not-dodgy but could-be-better-written, rather than agenda-driven bullshit.)

  17. HTML fail. paging Ben Goldacre. Oh, also there’s a more-or-less spurious passive smoking justification (which, if true, would be good on non-Mill grounds, but isn’t) – I’m talking about “things that are actually backed up by data”.

    Final point: it’s interesting that people don’t seem to give the same treatment to press releases about pub closures – which didn’t increase substantially from their previous trend rate after the smoking ban (although snooker/pool halls did, interestingly enough).

  18. JohnB: “The earlier studies that suggested 20%+ declines *were* revealed to be bollocks (and really obvious, stupid bollocks – the credibility of the researchers responsible for those ones, which is not the same as the team here – has to be in question). That’s why the commentary piece that Belinda highlights exists.”

    Yet the BBC still quote them, like in today’s article, for example, paragraph 11. This is the study rubbished by the BBC itself in Belinda’s link.

  19. “which didn’t increase substantially from their previous trend rate after the smoking ban”

    How do you work that one out?

    2004: 478
    2005: 102
    2006: 216
    2007: 1,409
    2008: 1,973
    2009: 2,365

  20. Can I please just be allowed to sit in a pub with my friends without them having to disappear outside for a fag?
    Let the “DoH Policy Research Programme ” fund anti-smoking studies and inform me of their results. But let me decide.
    I know johnb and other authoritarians are just trying to protect me from myself, and I should be flattered, but I just find it creepy and annoying.

  21. Dick: thanks & sorry. I was looking at on-trade (ie in pubs & clubs & bars & restaurants) alcohol sales data as a proxy, because that’s what I happened to have on hand from my last alcoholic drinks market report.

    It’s interesting that the rate of decline of on-trade alcohol consumption didn’t change, but that pub closures did rise – are we assuming the narrative here is that people who used to drink a pint or two in old-man pubs are now drinking at home, but this is being offset by feral yoofs drinking gallons of cocktails at chain city pubs?

    It’s also interesting that the rate stepped up in 2009, given that pub closures pre-06 were largely ascribed to property prices (ie if you’ve got a profitable pub that pays you a tolerable living, but you can sell it for flats for a million quid, you’d have to be seriously committed to pubcraft to not), and 07-08 to the smoking ban – but I can’t see how it could’ve been a factor in the 2009 closures.

    (one factor I guess is that on-trade data covers restaurants, whereas BBPA is pubs-only. I’m not sure it’s a terrible thing if people go to a restaurant and drink a bottle of wine with dinner instead of going to the pub and drinking a couple of pints with a pie, but maybe that’s just me…)

    Anyway. In the absence of detailed landlord and customer surveys covering closed pubs, breaking down “we can only afford to drink paraffin and live in cardboard boxes”, “changing tastes”, “landlords want the cash from flats”, “exodus of smokers”, and “supermarkets selling cheap booze” is going to be a complete voodoo exercise.

    Kit, you could try *learning to fucking read* before handing out personal abuse: I’ve said, on this page, “the public health justification for the smoking ban – which, incidentally, I don’t view as a legitimate justification on JS Mill grounds”…

  22. Johnb, I am sorry but I am glad you find being called authoritarian personally abusive.

    But tonight I was going to a “restaurant and drink a bottle of wine” but now I will have the image of johnb standing behind me nodding with approval. I’m sorry johnb but that’s creepy. 😉

  23. I am not a doctor so I might be talking bollocks, however……

    Is it really feasible that someone who smoked like a steam engine for 20 years, is reducing his risk of heart attack by smoking a couple less ciggies due to the smoking ban. It seems highly improbable to say the least.

    I alsways thought that smoking was a long term risk factor.

Leave a Reply

Name and email are required. Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.