Unhealthful News 206 – Does journalism school un-teach people how to count, "Shisha Clampdown" edition

H/t to @cjsnowdon for pointing out the BBC article reporting the oh-so-shocking turn of events, that bans on shisha cafes (hookah bars) have resulted in them going underground.  I mean, who would have ever thought that banning the social component of a popular legal activity might cause people to gather clandestinely instead?  Or that by banning something, the ability to properly regulate it — e.g., to minimize fire risk — would vanish?

But what struck me about the article was the claim,

The World Health Organization has advised that a 40-minute session on a waterpipe is the equivalent to the volume of smoke inhaled from at least 100 cigarettes.

Yet 84% of those surveyed thought it was fewer than 10.

Now maybe Sarah Sturdey (the author of the article, who deserves to be shamed by name for this) could have just transcribed that first bit from somewhere without engaging her brain at all — basic reporter behavior.  But you would think that immediately following it by a radically different belief would cause her to think “wait, do I believe that?”

The WHO claim is pure fiction.  It is an n-th generation telephone game miscopy of some claims that were basically made up from whole cloth, tied to a tiny bit of reality.  Snowdon wrote an extensive report about the apparent provenance of the disinformation, I have written a couple of brief comments about it, and there are plenty of other analyses out there too.  But let’s set aside the crazy option of Ms. Sturdey doing some research before writing her article, and just consider what would happen if she had a third-grader’s command of math and had thought about what she was writing during a quick trip to the restroom (or rather, I suppose, it would be the loo):

Hmm — “volume of smoke inhaled”.  To smoke 100 cigarettes in 40 minutes would be more than two per minute.  That is much faster than anyone smokes, and indeed would require more total inhaling then someone normally does in that much time, and for every breath to be smoke.  That is close to physically impossible.

She would then have time during her inevitable hand washing — she is a health reporter after all — to think:

Smoking even 10 cigarettes in 40 minutes would require about 1/5th of all breaths to be a solid drag.  Smoking that much that fast would be an extreme intense session, whatever was being smoked.    Gee, maybe “those surveyed” are not morons after all.

Notice that this requires no subject matter knowledge about the relative harm of a lung-full from a cigarette as compared to a hookah.  Someone could believe the common claim that hookah smoke is much worse for you (an absurd claim, given what we know about heat-not-burn smoking, and one not supported by any evidence), but still be able to do the simple “volume of smoke inhaled” arithmetic.

Frankly, I am a bit concerned about the 16% who thought that it was typical to smoke a greater volume than 10 cigarettes’ worth in 40 minutes.  But I guess that is why society has a place for people who are just really bad with numbers and are willing to make claims about matters they are clueless about, like taking a job as….

[You know, just forget it.  It is too easy a parting shot.  I will just let readers finish the sentence themselves if they want.]

Advertisements

12 responses to “Unhealthful News 206 – Does journalism school un-teach people how to count, "Shisha Clampdown" edition

  1. To be a fair the 'journalist' reported 100 cigs not 100 cigs in 40 minutes. This implies volume of inhaled cigarette smoke from 100 cigarettes is equivalent to the inhaled volume of shisha in 40 minutes.

    From http://www.tobaccofreeu.org/pdf/Hookah.pdf
    There is this claim….
    “MYTH #4: Smoking a hookah is not as addictive as smoking a cigarette because there is no
    nicotine.
    TRUTH #4: Just like regular tobacco, shisha contains nicotine. In fact, in a 60-minute hookah session, smokers are exposed to 100 to 200 times the volume of smoke inhaled from a single cigarette.(5)”

    This ref (5) is to an expert answer from Richard D. Hurt, M.D at the Mayo Clinic.
    http://www.mayoclinic.com/health/hookah/AN01265
    ” Hookah smokers may actually inhale more tobacco smoke than cigarette smokers do because of the large volume of smoke they inhale in one smoking session, which can last as long as 60 minutes.”

    No specific number of cigarettes is given and there is no further reference.

    The 100-200 version appears in a WHO advisory from 2005 at http://www.who.int/tobacco/global_interaction/tobreg/Waterpipe%20recommendation_Final.pdf

    “3. A typical 1-hour long waterpipe smoking session involve's inhaling 100-200 times the volume of smoke inhaled with a single cigarette (6)”

    Ref (6) is to a study. “Towards a topographical model of narghile water-pipe café smoking: a pilot study in a high socioeconomic status neighborhood of Beirut, Lebanon.” at http://www.ncbi.nlm.nih.gov/pubmed/15388286

    This claims “Results showed that the “average” water-pipe cafe smoking session consists of one hundred seventy-one 530-ml puffs of 2.6-s duration at a frequency of 2.8 puffs/min.” From the abstract there is no comparison to cigarette inhaled volume.

    In “Subjective and Objective Measurement of Cigarette Smoke Inhalation*”
    Martin I. Tobin, M.B.; Gilbert Jenouri, B.A.; and Marvin A. Sackner, M.D., F.C.C.P. at http://www.ncbi.nlm.nih.gov/pubmed/15388286 We have some inhalation volumes for cigarettes..

    “The mean volumes inhaled varied widely from 0.27 to 1.97 L, with a group mean (± SD) of 0.79 ±0.45 L. Mean duration of smoke inhalation varied from 2.0 to 6.8 seconds, with a group mean of 4.5 ± 1.3.”

    Inhaled Hookah = 2.8 puffs/min * 0.530l = 92.75l / session
    Inhaled 1 Cigarette = 0.45l (unless this is per puff in which case it is 4.5l/cig)
    volume sheisha inhaled compared to a single cigarette = 92.75 / 0.45 = 206.11 Oh! This is per session not 40 minutes though.

    Shisha for 40 mins –> 2.8 * 0.53 * 40 = 59.36l
    Inahaled cigarette 0.45l (or 4.5l/min 10 puffs) gives a comparison of 59.36/0.45 = 131.91 (13.19 for 10 puffs).
    Again within the 100-200 claim if the 0.45l/cig is correct although it does seem that is per puff so the comparison should be between 10-15 cigarettes per 40 min session.

    The numbers are confusing since the topo claim is 171 puffs per session at 2.8 puffs/min which seems to mean the 'session' lasted over 7 hours.

    For cigarettes it would appear that the volume / min is (0.45 * 10) 4.5l at 10 puffs a minute where as for the Hookah it is 0.53l/puff * 2.8 puffs/min = 1.484l / min.

    The cigarette study is also confusing as it includes smoke/air mix and ambiguity on whether it is per cigarette or per puff.

    I cannot find a study using 40 mins directly. There is mention of 30 mins, 60mins and session. The tobaccofreeeu claim is only supported by an expert's answer to a question without supporting evidence.

    The WHO claim does not seem to be supported by the study it references. The claim also seems to be a factor of 10 out.

    My maths may be inaccurate, there could be more up to date studies and also I am totally confused with the numbers! Oh Well, overall I think I prefer your explanation :)

  2. On the first “to be fair comment” — I agree there is no statement about how long was spent on the cigarettes, but that does not matter. The claim is that you can inhale that much volume in 40 minutes. So the logic is that if you are doing it with a hookah in 40 minutes then the same intensity of cigarette smoking in 40 minutes would be possible; but there is no way someone could smoke that many cigarettes in 40 minutes, and therefore they could not take that many drags on a hookah either. Unscrupulous “researchers” and bad journalists can trick readers into believing impossible things about weird exotic activities, so this just points out that the exotic activity is not physically different from the familiar activity, and thinking about the familiar demonstrates the impossibility.

    I will have to come back to the numbers with more attention than I have available right now. But since the logic I presented shows that the reporter's claim is obvious wrong, the same claim is no less obviously wrong when published somewhere else, and the “researchers” responsible for those claims are just as innumerate as the reporter. Also, keep in mind that a lung-full is a lung full, no matter what you are inhaling, so the volume per drag is going to be quite close to the same. FInally, the notion of “average” for something that is entirely behavioral makes it pretty clear that these researchers should have stuck with something easy like chemistry and avoided studying people.

  3. Correction: Not sure Inhaled Hookah = 2.8 puffs/min * 0.530l = 92.75l / session is correct, I think it should be Inhaled Hookah = 2.8 puffs/min * 0.530l –> 89.04 / 60 mins.

    I wanted to find the basis of the claim and . The claim is about the volume of smoke only so I tried to find out what that volume was. A lungful is a lungful which is where double inhalation, smoke first then air, and the actual volume of smoke comes in.

    I think the whole claim was designed to mislead but wanted to be assured it had no basis either in common sense or study. It does seems the claim lacks any substance.

  4. Hmm, no, sorry I'm with westcoast on this. I think whether it is constrained by time or not makes a difference to the claim (if you take it strictly – that the volume of smoke/air mixture inhaled from a shisha session could be many times the smoke/air mixture inhaled from a cigarette).

    It doesn't really matter to the claim if you can smoke 100 cigs in 40 minutes or not, different thing is different. If I loaned you my bong that burns horse manure and emits smoke to the user at 75°C , I'd imagine you'd be able to inhale less volume of smoke/air mixture from that in a given period of time than either cigarettes or a hookah. But it obviously wouldn't be because the human body simply couldn't inhale that volume of *anything*. There are factors like palatability, temperature, particle matter concentration, pH etc.

    It does seem that the average volume of smoke/air mixture inhaled from a cigarette is around 0.5 litres (e.g. http://jnci.oxfordjournals.org/content/92/2/106/T2.expansion.html), so in pure terms of volume of smoke/air mixture inhaled, the claim seems in the right area if in a shisha session each *single inhalation* is about half a litre (from west's ref).

    Of course this being a much larger volume of smoke/air mixture obviously doesn't necessarily equate to a directly proportional larger increase in risks for lots of other reasons, and it could be argued that volume isn't a particularly useful measure of exposure in this case anyway (cigarette smoke could be more particle dense etc – I honestly don't know). But it does seem to me that to tackle the claim you'd need to address those issues, not (what I think) is a red herring of 'but you can't smoke as many cigs in that time'.

  5. Ok, I guess one thing I should take from the comments is that I was a little hard on the reporter. What would have caused me to say “hey, wait a minute…” on a trip to the loo is not as obvious as I thought it was.

    Rory, I think part of the confusion might come from this “smoke/air mixture” construct, which might imply that volume of smoke is not the same as volume inhaled. The construct does not make sense, because smoke is a suspension in air of particulate matter from burning (definitions might be a bit more general or specific, but they all involve a suspension in air). So if you mix smoke and air you get… smoke. Just smoke. (At some point you might dilute it enough that you no longer think of it as smoke, but that is not relevant in the range we are talking about.) So volume of smoke is volume of inhalation — there is no dividing out the smoke from the air.

    So the *volume* of smoke inhaled from 100 cigarettes is what it is, regardless of how many days you stretch it out over. But since this is the unit that — the claim goes — is smoked in 40 minutes (a statement of rate: volume/time), it makes sense to think of smoking cigarettes at that rate as the reality check. Same volume, same time, more familiar activity. It is is utterly implausible to inhale the volume of smoke inhaled from 100 cigarettes in that time, then it is utterly implausible to inhale that much smoke from another device. Thinking about how fast you would have to be smoking cigarettes tells you that It would take some effort to inhale the volume of *air* equivalent to the smoke from 100 cigarettes in 40 minutes — you would need to be exercising.

    I will still get back to the numbers later. It might not be until Monday. So check back or look for the tweet.

  6. Fair enough, on reflection I can see why 'smoke/air' doesn't make much sense as I've written it, I guess the point was that not all smoke is equal in terms of other characteristics (particle concentration being one), some of which might facilitate consuming greater volumes more easily.

    I'd still think that (providing the measurements we are basing this on are any good at all) the volume of smoke from 100 cigarettes (say 0.5ltrs*100=50ltrs) isn't ridiculously implausible. E.g. according to the wiki, respiratory volume for a minute at rest is 5-8 litres for an average person(so 200 to 320 litres of air exchange in total over 40 min – doesn't seem that crazy to me providing wiki figure is in the right ballpark?)

  7. Ok, I think all of the confusion comes down to an error in one of the units and it is really pretty simple, as I originally claimed.

    Westcoast identified the problem, correcting for what may have been misreported or ambiguously reported in a paper, that cigarette inhalation volume is .45 l. That has to be per puff (it is in the right range, as opposed to the interpretation of it as being per cigarette), so of course it is about the same as a puff on a hookah. Frankly, I think that any calculation out there that is based on the assumption that puff volume is different is suspect — we need a lot more data on what is normal behavior for a particular population (more than just a single study of a single population) to make claims about that.

    So, if we work with the near isomorphism between puffs and volume, then a 171 puff hookah session must have about the same volume as 171 puffs of cigarette. As westcoast notes, that is about 17 cigarettes, not the absurd 100-200 that the (*cough*) “experts” claim. And that was based on a bit more than an hour (171 puffs at 2.8/min = 61 min), so we are down in the neighborhood of 11 cigarettes worth of puffing for 40 minutes. And finally, that is a pretty intense smoking session; no doubt you can find a population to study where this is “average”, but it came out to about 1/3 of all breaths being a hookah drag, without break, for more than an hour (and implicitly assumes that the hookah was kept full and hot the whole time, and that the user was not just sucking slightly flavored air out of habit half the time).

    So we are talking a very intense smoking session getting the volume in the order of 10 cigarettes. Clearly increasing that by a factor of 10 would be physically impossible. Roughly speaking, every breath would have to be smoking, and respiration would have to be increased by a factor of three compared to resting. Completely implausible. No reprieve from the accusation of press innumeracy there.

    Even if a hookah puff is 10% or 20% more volume, on average, than a cigarette puff, that does not come close to making up for the error. And taking a deeper breath would need to be averaged out across a minute or we are back to the increase in respiration (need to exercising or will get hyperoxic — hmm, I guess high CO saturation could help avoid the hyperoxia, so maybe it is possible to breathe like you are jogging while smoking a hookah, though I think such behavior is pretty rare).

    If anti-tobacco people were honest and numerate, we might see some advice like “if you are going to smoke hookahs, keep it down to one puff every minute or two, or maybe take some time out after 20 minutes — if that is what you are used to, you might find it just as satisfying as having the hose welded to your mouth. Of course, then they might have to recognize that allowing the serving of tea and biscuits at the hookah bar would help encourage that. But since we are already past the “no food can be served there” step on the road to prohibition, they are stuck with it, and the smoker is left with nothing else to do with his hands.

    Anyway someone must have done a complete takedown of the myths, maybe Chaouachi. I wonder if the lie/error about volume that is an order of magnitude higher than is even plausible traces to that one paper, which apparently implies that an entire cigarette is consumed in half a liter of inhalation (one breath). I am also curious about the lack of any study or calculation that uses 40 minutes as the period, given that claims about that period are made.

  8. Thanks for persevering and coming to some reasonable conclusions. When I think of Hookah pipes, plumes of smoke come to mind. This makes it easy for this claim to pass without either the reasonableness test Carl applied or the issues Rory raised.

    As for the Journalist, well they will just say they were reporting what the experts say. In this case the 'expert' is the WHO. We have found this claim maybe the result of an error or unreasonable assumptions and can, when the chance or need arises, challenge it. So overall a useful discussion.

  9. “…that cigarette inhalation volume is .45 l. That has to be per puff (it is in the right range, as opposed to the interpretation of it as being per cigarette), so of course it is about the same as a puff on a hookah.”

    Er, well it only has to be per puff if you are trying to make it fit with your original figures!

    A quick search finds that that figures in that range are definitely per cigarette. With each puff being in the 30 – 60ml range.

    e.g. here from 1981 (pdf)

    from 2009 in adolescents

    from 1999 in adults

    and Imperial Tobacco use similar assumptions (pdf)

    So it's more than one paper in a single population, and yes, there are going to be variations by individual population characteristics, but it still seems a pretty consistent picture building up. I think the burden of proof would be on you to demonstrate people puffing on cigarettes with 0.5 litre puffs.

    Why I'm becoming a bore about this now is because it has reasonable ecological validity to me. I've both smoked cigarettes and used a waterpipe in the past. With cigarettes I used to take a contained puff into the mouth then draw it into the lungs. With a waterpipe it's a big suck straight into the lungs with no stopover of smoke in the mouth. I recall pretty clearly there's a giant difference in inhalation required, so a order of magnitude difference doesn't seem unreasonable if what applied to me applies to at least some other people (and everything I've read so far in a brief search makes that seem likely).

    Again, doesn't say much about harm … but the arithmetic is against you I think.

  10. Hmm, fair enough. So I guess the problem is that I should have stuck with that smoke-air mixture concept you offered rather than trying to simplify to just “its all smoke”. We need that concept — which is to say, paying attention to the dilution of the suspended particulates — in order to make a legitimate comparison. So the upshot is that if you think of the cigarette puff as just the bit when you are drawing through the cigarette, and not including the rest of the breath that most smokers take to inhale it or already have in their lungs and exhale with it, then it is much smaller than a breath.

    (Aside: It is pretty clear that stopping with the mouthful — cigar style — would be a major harm reduction, but no one pays much attention to that.)

    By that definition, the claim can perhaps be promoted(???) from being literally false to being a lie that can be interpreted in a way that makes it literally true. The concentration of the smoke from a hookah lungful is much more comparable to that from the whole diluted lungful that includes a cigarette puff, not the concentrated mouthful that exist at the moment of puffing. Or thought of another way: it should be obvious to anyone who has been around both that the thickness of the smoke in the room after someone smokes a hookah for an hour is going to be much closer to that from 10-20 cigarettes smoked in the same space in the same time, not 100-200. But since there is no clearly defined right way to make the comparison, it is possible to make a comparison that violates no claims or norms but still is obviously extremely misleading based on the way most people would read it — a lie, in other words.

    ….

  11. ….

    By the logic that makes the claim literally true, you could claim that being in an indoor space where smoking is taking place caused you to inhale the same volume of smoke in an hour or two as smoking 1000 cigarettes. Yes, exposure to second-hand smoke has even more smoking volume than hookah smoking — every breath you take is smoke. So if you limit the smoke volume of the cigarette to the highly concentrated puff, you get a volume ratio in that range. The same is true if you are breathing around a wood fire or just in a highly polluted city. On the other hand, if you then recognize that the cigarette and hookah smoker are both being exposed to their own ETS for the whole session, then it has to be a tie. Obviously, concentration/dilution matter, so I need to focus on that to make this case.

    So, what does that say about the reporting? I am not sure it offers any relief. I might be wrong, but I suspect few reporters who parrot the claim have ever gone down this path of figuring out, “it sounds absurd, but ok, I see how it can be interpreted to be literally true”. If they had, presumably they would at least throw in a perfunctory “but the cigarette smoke is much more concentrated” or something. In some sense, it is the reporter's job to make it so that I do not have to figure out (thanks to the contributions of two helpful readers, without whom I would not have done so) in what tortured sense a particular claim is true. So what exactly were those survey respondents asked and what did they think they were answering? If people's response to a question about a reasonably common phenomenon (i.e., it is not about some obscure budget figure, tiny probability, or astrophysics — something we would expect most people to not guess well at) is generally off from the “official” answer by more than an order of magnitude, it is worth asking why. In this case, I would guess the majority were doing what I did — thinking in terms of the most equivalent version of “the smoke” without it even occurring to them that a different way of measuring might be used.

    So the reporter should still say “wait a minute, what is going on here” and perhaps take the bold step of asking those feeding her the information “what do you mean by 'volume of smoke'”? At least it might salvage that 16%, who might have correctly been doing the comparison in the misleading — but not wrong in any clear sense — way.

  12. Ah, right, see where that's coming from now, makes sense, thanks. Hadn't correctly put together the concentration issues with the volume measurement stuff.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s