Opinion on Harvard study

[VNeil]

....

1. In 2011, National Institute for Occupational Safety and Health proposed a recommended short-term exposure limit of 25 parts per billion (ppb) and a time weighted average exposure of 5 ppb. How do these recommendations correlate to what was found in the study?

This gentleman attempted to convert ppm of exposure (via breathing ambient air) into micrograms of diacetyl (or could be done for any other compound) and I think he did a pretty good job...

Everybody is talking about vaping and "popcorn lung" again, so here's a graph

Using his amended formula (which is slightly different than the formula on the site at the moment), for 5 ppb over an 8 hour day I calculated 69 micrograms. For the short term exposure I calculated 345 micrograms (about equal to a single cigarette).

ETA: I scratched the calculation for ST exposure because I have no idea what that would mean in a vaping context. The 345 micrograms is wrong because it is based on an 8 hour exposure, which is certainly not what NIOSH intended. So I would stick with the time weighted exposure and call it a day

If you try to use his formula, replace the 0.99 g/ml in the formula with 0.0036 g/ml (or pick your density for diaceytl in ambient air, as was discussed in the comments).

If applied to vaping, I would expect the time weighted 5 ppb exposure to be used, suggesting you would want 69 micrograms or less in the ejuice you vape over an entire day, in order to comply. If you vape 10ml per day, that would mean you would want 0.69 micrograms/ml or less. Which is basically at the limits of detectability. I believe that level (0.69/ml) is about 1/10th what most people here would consider "trace and meaningless", just to try to put context on it, but everyone's definition of "trace" might and will vary.

 

[Falconeer]

The , or should I have posted THE THING, about this Popcorn tale is that I have seen the destruction it can cause ( and my wife is a highly quaified nurse/ depute matron etc and I was a Disablement Advisory Services Manager prior to retirement; so we feel we are well enough qualified to offer an assessment) -three years ago in Spain where we live for half of the year there was a vape shop on every corner ( now, tobacconists in Spain or Estancos as they are called are at the pleasure of the Government who sets the prices twice a year and rakes in the taxes, and who therefore have a vested interest) made a point of publicising this tale of woe big style. Now you have to hunt to find one vape shop in a city. That's the damage this misinformation can cause. Most Spanish folk who transited to Vaping have now gone back to smoking - or at least in our village they have.

Misinformation? "Popcorn lung can only be proved/ established AFTER A POST MORTEM. Go on Google it if you don't believe me.

 

[David Wolf]

I am in no way an expert in the field, I just have some experience in reading scientific papers…
Article itself is here: http://ehp.niehs.nih.gov/wp-content/uploads/advpub/2015/12/ehp.1510185.acco.pdf


Short version:


Again, I am not an expert in methods of measurements which were used, so let’s for now accept results.

Bad news:

One of the cigalikes tested in the study provided 528 um (microgram) of diacetyl (D). For one real cigarette high amount of D is equal to 433 [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4137810/]

One cigalike contains about 1 ml of juice (correct me if I am wrong). So, 1 ml of that juice is equal to 1 cigarette in terms of amount of diacetyl. It means that if a vaper is vaping 20 ml of this juice a day[*], vaper gets as much D as from 1 pack of cigarettes with high amount of D. So, yes, vaping with specially selected juices can provide as much diacetyl as smoking.
* I mean not 20 cigalikes a day, but 20 ml of juice with the same flavorings but lower nic.

Good news:

Just two juices were really bad, close to cigarettes (both of the same company). Couple of other juices were somewhat bad, and about 34 (of 51 tested) juices were at least 100 times better than the worst one.


Authors attitude:

Authors stretched their method a lot to get barely detectable amounts of D. Their agenda was very visible: to show as much bad things about vaping as possible. But anyway we should be thankful for results they gave us – we should know what we vape. Not a lot of measurements (with whatever agenda) were made to date.


Some details.


It looks like authors have not much knowledge of e-cigs. They made measurements on cigalikes never explicitly stating it.


Method of obtaining vapor: they imitated 8 seconds draws whith 15-30 seconds resting time between draws. Looks OK, but they continued test to the full exhaustion of cartruges, determined by the lack of visible emissions in the chamber. So dry hit for the last draw (at least for one draw). Good they did not measure formaldehyde…


Measurements, especially of lower quantities, are quite questionable. Authors were too eager to measure things they cannot reliably measure. Some juices were measured multiple times and results were disastrous in low D concentration range. For example 4 measurements of the same cigalikes gave results 27.0, 1.6, 0, 0. Authors try to explain these variations by non-consistent manufacturing process, but much more probable their method is not suited for low concentration measurements.


Disclaimer: I have no personal interest in knowing of levels of diacetyl in flavorings. I DIY and use just 0.5-2% of flavorings (instead of usual 20%). I consider myself safe with diacetyl. Anyway, diacetyl is potential risk factor (not really well established) and everybody should decide for himself what is acceptable and what is not. To decide we should know how much (if any) of diacetyl is in juice/flavor.

Good work, I think you've provided a much fairer assessment of the actual data and test than the study authors. I do believe though, that you've added your own spin to the tests regarding the "Bad News" by extrapolating a"what if" scenario that had nothing to do with the actual cigarette cartridge sold and tested. Brand C "Peach Snaps" was bought in the form of a cartridge, and testing indicated Diacetyl=238.9 (not 528) and Acetoin= 529.2. Purchased in cartridge packs, typical usage would be one to two cartridges a day at fairly high nic levels. Assuming your 1ml per cartridge and extrapolating it to 20 ml for use in a tank would require a lot of money and time squeezing all of that juice out, or you have to make the assumption that the manufacturer also supplies the exact same formulation for tanks - a big unknowing assumption, it might be that cartridge formulas are stronger, or that they only sell it in cartridges, etc. Some ejuice suppliers sell different formulations for drippers and tanks, and it would not surprise me to find different formulations for cartridges. In any case, we don't even know if the supplier sells that flavor in equice form, or what that formulation is if they did. You would have to test the ejuice, not the cartridges.

Which, back to the actual tests and away from conjecture, the study did test some e-liquid- "In addition, we evaluated 11 e-liquid flavors that are inserted into a cartomizer (disposable cartridge and atomizer system) (Brands H and I, 6 and 5 flavors, respectively)." So assuming that Brand H and I are sold for tank formulas and not just cartridges, your comparisons for vaping high quantities of e-liquid would appear to be on more solid ground (with the assumption as noted) if you used them on your example.

Now on the the study itself, a more thorough examination (thanks to your good example), I find that several of the flavors were repeated in testing, from the same pack per the study, with significantly different results. This is rather disturbing - either the vendors can't maintain consistency of formulation within the same pack or something is amiss in the testing. Here's a worst case example, but there are others:
Flavor Brand Flavor Type Diacetyl (2,3-butanedione) 2,3-pentanedione Acetoin
Pina Colada C Cocktail 1.6 <LOD 130
Pina Colada C Cocktail <LOD <LOD <LOD

The flavor above was also tested two more times, see the actual study table 2 for the varied results.
http://ehp.niehs.nih.gov/wp-content/uploads/advpub/2015/12/ehp.1510185.acco.pdf
Disclaimer: I vape.