Talk:Electronic cigarette

Glanze Redux

Just listened to an interesting 5 mins or so feature on Glanze's recent paper (see above) on today's More or Less (radio programme), a BBC radio programme covering statistics etc. Some of you should be able to pick it up online from the website linked at the article, starts about 5 mins in. Linda Bauld, Ann McNeill, Peter Hajek (co-author of the Cochrane Review - "grossly misleading" he says), Robert West ("mashed together very different studies") - all saying the paper should not have been published, as did an un-named person who peer-reviewed it for the Lancet sub-journal. Glanze & the editor unrepentent. Johnbod (talk) 16:52, 5 February 2016 (UTC)

That's what you get when you mix activism with science. I'm sure it will be straightened out in coming reviews.--TMCk (talk) 18:11, 5 February 2016 (UTC)

Well from a WP stance, as i mentioned earlier[1] its simple: The Glantz paper is a secondary WP:MEDRS review, thus reliable and can be used. The More or Less program is interesting - but cannot really be used. If the paper is flawed then we as WP editors must wait until it gets resolved in the peer-reviewed press. --Kim D. Petersen 09:39, 6 February 2016 (UTC)

Well, the American Legacy Foundation (currently Truth Initiative), which is the largest anti-smoking organisation in the USA, has publicized a systematic review ("The findings were compiled from an in-progress comprehensive systematic review of all published scientific literature on e-cigarettes conducted via a PubMed search through February 19, 2015") which criticizes a pre-published version of Glantz's meta-analysis. Can their findings be reflected in WP?GreyZig 14:37, 16 February 2016 (UTC) — Preceding unsigned comment added by Zvi Zig (talk • contribs)

Why all the refs in the lead?

The lead section of a Wikipedia article is supposed to summarize the article's contents, so inline citations (which make it look like the material was added to the lead first) should be kept to a minimum. And yet somehow this article's lead has over forty!? Also, the lead is enormous... just saying...

Hijiri 88 (聖やや) 09:58, 22 February 2016 (UTC)

Most leads on medical articles are referenced, and should be. Most leads are also much too short, and this is a long article on a complex topic. The lead now has 5 paras, one more than WP:LEAD suggests, but personally I think that's ok. Johnbod (talk) 12:45, 22 February 2016 (UTC)

Can you cite tge PAG that says "Most leads on medical articles should be referenced"? This seems counter-intuitive to me. The body should be thoroughly referenced, but the lead should comprise a summary of the body. It should be based on the content of the article itself, not on external reliable sources. Hijiri 88 (聖やや) 13:58, 22 February 2016 (UTC)

Can you provide a PAG that says that the lede should not be referenced? AlbinoFerret 14:08, 22 February 2016 (UTC)

WP:LEDE states that ”[t]he lead serves as ... a summary of [the article's] most important contents”. It does say that the lead's contents should be verifiable (a given), but this in my experience more usually interpreted as meaning that the lead should summarize the sourced contents of the article body. The only things it says need citations are contentious statements. The enormous number of citations in this article's lead (some of them not cited anywhere else in the article!) implies it was composed independently of the rest of the article, which would be a clear violation of the project norm. Hijiri 88 (聖やや) 22:59, 22 February 2016 (UTC)

I take that as a no, you do not have a PAG that says the lede should not be referenced. AlbinoFerret 23:16, 22 February 2016 (UTC)

The relevant section is WP:LEADCITE, which doesn't prohibit going overboard with cites (that's left to editorial judgement). Common practice is to avoid citation clutter in the lead as, per MOS:INTRO, it is "a concise version of the article" meant to "briefly summarize the most important points covered in an article" (where the cites are mandatory and thus redundant in the lead). Cites in the lead interfere with its readability, particularly when they are numerous. This lead is a particularly ugly example, but I won't comment on whether it's justified. Curly Turkey 🍁 ¡gobble! 23:46, 22 February 2016 (UTC)

@AlbinoFerret: So let's get this straight: I post an OP comment accurately summarizing the PAG that says that leads should summarize body contents, and another user shows up and makes a bizarre claim that articles on medical topics usually have more sources in the lead than other articles. This "PAG" does not appear anywhere on-wiki. When I ask the user to quote the PAG, you show up and dodge the question, insisting that I quote the PAG that I already quoted. I do this, and then you promptly ignore the PAG that I quoted at you. But still no Wikipedia PAG has been provided that says "Most leads on medical articles should be referenced". MOS:MED actually says the opposite: Adding sources to the lead is a reasonable practice but not required as long as the text in question is supported in the body of the article. So why are there some sources cited in the lead that aren't cited in the body? Does the material they support in the lead not appear in the body? If so, that is an explicit violation of the core principle of WP:LEDE that The lead should ... summarize the body of the article with appropriate weight.. Hijiri 88 (聖やや) 04:49, 23 February 2016 (UTC)

To say, as I did, "Most leads on medical articles are referenced" will only seem a "bizarre claim" to someone who has never looked at any of our developed medical articles. Johnbod (talk) 13:55, 23 February 2016 (UTC)

There are statements in the lead that aren't cited in the body? That's a straight-up violation. Are you sure they weren't in today's addition (the one that added a fifth paragraph) that has since been reverted? Curly Turkey 🍁 ¡gobble! 05:06, 23 February 2016 (UTC)

I can respond to Hijiri88 by stating that no PAG supports this either way, and LEADCITE allows for a fully cited lede. That medical articles normally have fully cited ledes doesn't mean any PAG requires them to have citations, but that the community of editors has come to realize that it is very helpful—it may at some point be added to a PAG, but for now that hasn't been done as it is not required. WP:MEDRS gives an indication as to why medical statements are sensitive, not only because of their impact on lives, but also because of the controversy surrounding any single statement.

Now, this may be going to in depth for this discussion, but most of our readers (especially on medical articles) do not read past the lede, and having sources present there is very useful for that group. (I will not provide data for this here, but it exists). Also worthy of notice is how certain editors and even IPs will tag any statement without a directly visible source in medical articles with a {{cn}}-tag, regardless if it is in the lede or in the body.

So, lets move this discussion from the unproductive to the productive. We identified that the prior 5-paragraph lede had 6 citations for a single statement – which is CITATIONOVERKILL – and normally a sign of dubious statements. I can not see any similar issues with the current content — does anyone else? CFCF 💌 📧 07:28, 23 February 2016 (UTC)

There are statements in the lead that are cited to sources that are only cited once, in the lead. Admittedly, on a brief inspection the one I found was attached to a sentence that also cited two other sources, which means either that it is redundant or that the sentence includes information taken from all three sources, and some of that information does not appear in the body of the article. I don't have a serious problem with a fully-cited lead (though I think it is ugly and gives the wrong impression, and would never do it myself -- this attitude is not required by PAG, but it is encouraged), but unique citations in the lead are definitely a problem, as they are either redundant and give the wrong impression, or they are being used to include information in the lead that does not appear in the article proper. A slightly more thorough examination revealed that there are at present three such sources (currently numbered 15, 29 and 37); of these, only the last is not accompanied by other citations, and appears to be used to support the statement that "In the United States e-cigarettes are used by a significant percentage of ... adults". ("e-cigarettes are used by a significant percentage of youth" is attributed to another source that is used in the following section for "up to 13% of American high school students had used them at least once in the last month", but this difference in wording is not really a problem.) I don't have time to read the entire article at the moment, much less the tens of thousands of words of talk page archives, but has this sentence in the lead been discussed? Does it accurately and duly summarize some point that is made in the article? Summarizing "3.4% of American adults" as "significant" seems dubious... Hijiri 88 (聖やや) 08:12, 23 February 2016 (UTC)

The more controversial an article subject is, the more citations you will see in the lede. That follows per WP:V. If someone could question the statement, then it must be cited. --Kim D. Petersen 21:47, 22 February 2016 (UTC)

The more controversial an article subject is, the more citations you will see in the lede—not true. Richard Nixon (a Featured Article) is the subject of considerable controversy, yet there isn't a cite in his entire lengthy lead (even, surprisingly, for the statement "he is historically ranked as among the worst U.S. presidents"). Citations are allowed in the lead for particularly contentious individual statements, but never are they required in the lead, as any statement in the lead must be in the body, where it must be cited. Curly Turkey 🍁 ¡gobble! 23:50, 22 February 2016 (UTC)

• I tend to agree with the OP that the Lead does have a case of WP:Citation overkill. The 1st sentence of paragraph 2 has 6 citations! As a side note, I'm not sure when it grew a 5th paragraph. It only had 4 the last time I checked. ~Awilley (talk) 01:32, 23 February 2016 (UTC)

This poorly constructed and undiscussed edit inserted the 5th paragraph just a few hours ago: https://en.wikipedia.org/w/index.php?title=Electronic_cigarette&diff=706352255&oldid=706329637 CFCF 💌 📧 01:45, 23 February 2016 (UTC)

(edit conflict)Pinging User:Zvi Zig who seems to have added the 5th paragraph just today [2] ~Awilley (talk) 01:46, 23 February 2016 (UTC)

Have restored it to 4 paragraphs. Refs in the lead are fine. Doc James (talk · contribs · email) 02:31, 23 February 2016 (UTC)

Sorry I've not known the 4 paragraph lede rule...

Unfortunately, my edits have been reverted by talk. The concerns that I tried addressing in my reverted edit were...

1. The opening lede statement on health asserting all-embracing uncertainty is redundant vague. The direct statements on health risks and cessation, convey their respective degrees of certainty.

2. As a review by Cancer Research UK states, that "there is a consensus that e-cigarettes are almost certainly much safer than tobacco smoking". A wide range of sources concur on this point, as I have referenced in my edit which talk reverted. The phrase, "e-cigarettes are probably safer", doesn't reflect this point.

3. 95% less harmful estimate by Public Health England and others should be reflected in the lede -- "safer" is not quantified. The Pisinger review conclusion that "no firm conclusions can be reached..." should also be reflected, for balance. ZviZig/GreyZig 12:03, 23 February 2016 (UTC)

You are missing the numerous less than positive reports, it has taken time to find the balance that now exists between positive and negative sources. CFCF 💌 📧 13:41, 23 February 2016 (UTC)

• I've started Wikipedia_talk:WikiProject_Medicine#.22Adding_sources_to_the_lead_is_a_reasonable_practice_but_not_required_as_long_as_the_text_in_question_is_supported_in_the_body_of_the_article.22 - if we don't have a "PAG" (?!) explicitly stating what we normally do in medical articles, then we should. Johnbod (talk) 14:57, 23 February 2016 (UTC)

The original premise in this discussion is flawed. Our manual of style gives this guidance on the lead: "The lead serves as an introduction to the article and a summary of its most important contents." and "It should identify the topic, establish context, explain why the topic is notable, and summarize the most important points, including any prominent controversies." (my emphases), which is not the same as assuming that it must be solely a summary of the rest of the article and nothing else. Not every topic can be properly introduced through sources that already exist in the rest of the article, and that may be especially true of articles dealing with controversial topics. It is perfectly reasonable for a source that gives a broad introduction to a topic to be a good choice for referencing part of the lead, and yet not be sufficiently detailed to act as a good choice as a citation within a complex discussion later in the article, where more specialised sources are more appropriate. These sort of judgements belong to discussion in individual articles and it is necessary to seek a consensus in these cases. Apart from direct quotations, there's no rule that dictates whether citations appear in the lead or not; nor is there a rule that proscribes the use of a citation in the lead that does not appear elsewhere. --RexxS (talk) 18:53, 23 February 2016 (UTC)

You're not seriously suggesting the lead can be loaded up with information not in the body, are you? Curly Turkey 🍁 ¡gobble! 23:12, 23 February 2016 (UTC)

Are you seriously suggesting that the lead must only contain information in the body? I've linked the PAG pages that show how you have failed to understand that the lead performs more functions than just summary of the body. You are simply wrong when making false claims like "any statement in the lead must be in the body" - which is patent nonsense. The balance between the functions of introduction, establishing notability, providing context, and summarising the body is a matter for consensus between editors at the article, not by dictat from a non-existent policy. --RexxS (talk) 23:47, 24 February 2016 (UTC)

I'm seriously suggesting that serious justification is required for anything in the lead not appearing in the body. How do you justify keeping such information out of the body if it's important enough to be in the lead? How does that not violate WP:WEIGHT, for one instance? Curly Turkey 🍁 ¡gobble! 02:06, 25 February 2016 (UTC)

• Hijiri88 - please read WP:CITELEAD. It says that if anything in the lead is challenged, editors should provide in-line citations. You may not be aware that almost every line of this article has been fought over; there is nothing that hasn't been challenged. Because of that, I hope you can see that it is completely appropriate to have citations in the lead. In my view, the lead is probably about the right length as this is a long article. It could possibly be condensed more but doing so will probably be very costly in terms of volunteer time to negotiate that. There is nothing alarming in the lead that I can see, and nothing to really argue about. This has nothing to do with MEDMOS - that is a red herring. It is plain old WP:LEAD at work here. Jytdog (talk) 19:52, 23 February 2016 (UTC)

I am indeed aware of the controversy surrounding the article. I even discussed it with Curly Turkey some time ago, and the latter user pointed out how much more controversial this article apparently is on Wikipedia than, say, the State of Palestine. It frankly baffles me. But this article's lead currently appears to contain information not found in the body, or at best a dubious summary of what's in the body (3.4% became ”a significant percentage”, which is WP:WEASEL at best). This can happen as a good faith mistake when the lead is independently of the body, rather than based directly on the body. Any time the lead cites sources not cited in the body it should be a cause for concern. Hijiri 88 (聖やや) 06:05, 24 February 2016 (UTC)

I agree that nothing should be in the lead that is not the body. Why not try to just fix it (as conservatively as you can!!) and see how that goes? I would suggest that you understand what is driving the dispute before you get involved so that you don't walk into propeller blades. I'll drop a note on your talk page. Jytdog (talk) 08:35, 24 February 2016 (UTC)

Break

There is general agreement that Lead sections can have citations (nobody has argued otherwise) and that this particular Lead section should have citations. The problem that seems to be getting ignored in some of the arguments above is that this Lead section has too many citations. For instance, we have the following completely uncontroversial sentence with 4 (!) citations.

The e-liquids used in e-cigarettes usually contains a mix of propylene glycol, glycerin, nicotine, and flavorings.^{[17][19][23][24]}

There is also this beast with 7 citations:

The vapor can contain small amounts of toxins, including traces of heavy metals detected at levels permissible in inhalation medicines,^[16][27][6] and some potentially harmful chemicals not found in tobacco smoke^[28] at levels permissible by workplace safety standards.^[27][29][30]

I think the best way forward would be for somebody to judiciously prune some of these citations so we have no more than about 1-2 per sentence, and certainly no more than 1 per clause. If we find long sentences like my second example that can't be sourced to 1-2 sources it might be a good reason to break them up or look for possible WP:SYNTH. I could do this, but I don't know the literature as well as most of you regulars here so it would take me 5x as long. Volunteers? ~Awilley (talk) 19:46, 24 February 2016 (UTC)

That may be possible, but some sentences have been combined and the citations moved to the end. In that case it may be necessary to keep them. But the main components of ejuice is probably non controversial. AlbinoFerret 20:52, 24 February 2016 (UTC)

You don't need to remove cites; you can bundle them. That still leaves the issue of material in the lead that's not in the body, though. Curly Turkey 🍁 ¡gobble! 21:19, 24 February 2016 (UTC)

I now removed some extra references relevant to the sentence I edited on heavy metals. I also left only most authoritative sources in the sentence on e-juice (Cochrane and AACR reviews).GreyZig 21:46, 24 February 2016 (UTC)

Bundling is a good idea, but I'm not sure if it's compatible with the current reference system of naming the references and re-using them in the article. Reference 6 (Grana 2014) is currently being cited in 26 different sentences throughout the article. Bundling all the instances of multiple citations could lead to a lot of redundancy. Another option might be to convert the references using {{Harvard citation}} with short citations that link to the full citations in a References section, but that's a lot of work. ~Awilley (talk) 22:33, 24 February 2016 (UTC)

Well, if someone wants to do it, it would save a lot of headaches in the long run and make the article more pleasant to read. Still, that leaves the issue of material in the lead that's not in the body. Curly Turkey 🍁 ¡gobble! 22:59, 24 February 2016 (UTC)

I think that bundling is the way to go, but we need to check the citations for redundancy, there are quite a few that are essentially saying the same, do not directly support the statement, or just point to an article/publication that "talks about the subject".

Once example of redundancy

Non-smokers who use e-cigarettes risk becoming addicted to nicotine,[21] a chemical with a range of harmful effects.[22]

Both of the citations tend to support the first part of the sentence; but none of them (specially 22) support/talk the second part (harmfull effects of nicotine). While there is a full article devoted to the second part which in my opinion is better to link. Marianol (talk) 16:54, 10 March 2016 (UTC)

Repeating that there is material in the lede that doesn't exist in the body is all good—but you haven't actually presented anything that argues for this. Like RexxS says — we don't necessarily need the same sources, but for the lede to be a summary of the article's major points. The only substantive argument in this entire mass of verbiage is that 3.4% of the adult American population isn't significant — but since that includes nearly 10 million people it damn well is significant. Now I'm going to paraphrase Jytdog (though somewhat less elegantly) and ask you and Hijiri 88 to either come with constructive suggestions or stop wasting everyone's time on petty semantics and nit-picky interpretation of policy. CFCF 💌 📧 23:05, 24 February 2016 (UTC)

I'm not going to make assumptions and insult your level of education and/or choice of major like you did to me below, but "the American population "and "American adults" are radically different numbers. If what you are saying is that when the article body says "adults (as opposed to youths)" it should say "population" and the relevant line in the body is already misquoting the source, then the problem is even worse than I thought, and adults aren't mentioned anywhere in the body so whether 3.4% is significant would be a moot point. Hijiri 88 (聖やや) 04:43, 25 February 2016 (UTC)

Sorry. The above was a response to my misreading of CFCF's comment. I tend to say "American adult population" rather than "adult American population", and because I was reading the code as a diff rather than the actual text, with screen size and everything. Basically I messed up. It still doesn't justify him calling me "the troll", insulting my intelligence and level of education, and generally taking a dismissive and hypocritical attitude toward my presence on this page -- over an hour earlier than I made the above mistake. But an error is an error, and I apologize for making this one. Hijiri 88 (聖やや) 08:02, 26 February 2016 (UTC)

It looks like the 3.4% number comes from 2011. Later in the body we have the statement that "In the US, as of 2014, 12.6% of adults have used an e-cigarette at least once and about 3.7% are still using them." The source that is being used to say "significant percentage" in the Lead is a 2015 Reuters poll that says "About 10 percent of U.S. adults now vape, according to the online Reuters/Ipsos poll of 5,679 Americans conducted between May 19 and June 4. That's almost four times higher than a U.S. government estimate that 2.6 percent of adults used e-cigarettes in 2013." I think it would be best if we treated the Reuters article like a "primary" source and used numbers from secondary sources analyzing lots of polls. ~Awilley (talk) 16:35, 25 February 2016 (UTC)

Primary and secondary sources are not the concern. We should not be giving information in the lead that contradicts that in the body, even if both are reliably sourced. If the 3.4% figure in the body is out of date, the body should be updated first and then the lead should be edited to summarize what is in the body. The lead should never have been sourced independently of the body. Hijiri 88 (聖やや) 23:00, 25 February 2016 (UTC)

but you haven't actually presented anything that argues for this—you're right, but my point was that the issue was raised but hadn't been addressed. If there isn't material in the lead that isn't in the body, then somebody should say so.

but since that includes nearly 10 million people it damn well is significant—you have an empirical measure of "significant"? Curly Turkey 🍁 ¡gobble! 23:24, 24 February 2016 (UTC)

On that note, I'm still trying to find the place in the body that says "a significant percentage of American adults"... Hijiri 88 (聖やや) 01:37, 25 February 2016 (UTC)

So you're arguing that we should consider random musings of unengaged individuals? If you aren't going to take an effort to understand the underlying issues I am going to ignore you from now on, and I suggest others do so as well—per WP:Don't feed the troll(s). As for having "an empirical measure" of significant I suggest you brush up on what linguistics is, and why words are by definition intangible. CFCF 💌 📧 03:17, 25 February 2016 (UTC)

You've put on display more levels of ignorance with this statement than I have the energy to deal with. Curly Turkey 🍁 ¡gobble! 03:58, 25 February 2016 (UTC)

I find it incredibly ironic that at least two users have advised me that by posting on this page I am stepping into a minefield, but several users are also apparently criticizing me for not directly implementing my proposals without getting talk page consensus first. Actually Jytdog's was not so much a criticism as a friendly suggestion that I make a move that might get me in trouble in the short run. The last week has taught me to be slow to listen to the well-intentioned advice of friends -- I almost got blocked for it. I'd much rather follow my gut and discuss on the talk page first.

How is my pointing out that the lead of this article contains information not given in the article body, sourced to references not cited in the article body, and that this is a violation of Wikipedia standards, a "random musing of an unengaged individual"?

Hijiri 88 (聖やや) 04:25, 25 February 2016 (UTC)

1) Making one or two conservative fixes and then asking on talk, how was that, or 2) offering concrete proposals on talk for discussion, are much more productive ways to move things forward. i hope you didn't think i was advising you to go hog wild. :) Jytdog (talk) 04:29, 25 February 2016 (UTC)

Well, we can see what a minefield this is—Hijiri hasn't actually made any edits to the article, nor has he volunteered to, but he's already been templated and called a troll. Curly Turkey 🍁 ¡gobble! 04:35, 25 February 2016 (UTC)

(edit conflict) That's explained in detail at WP:OWN. Curly Turkey 🍁 ¡gobble! 04:30, 25 February 2016 (UTC)

ay caramba. providing a new editor to the discussion with formal notice of DS is not "templating", it is good process, and required for the DS to function. Curly Turkey. (which reminds me that I should make sure you have been notified) And I meant what I said - if you review these talk pages they are full of long, long, general conversations. On controversial topics it is way more productive to talk simply and clearly about concrete content. Make a conservative change and ask for reactions, or propose conservative changes. That's all. Now really, what do you want to change in the content, Curly Turkey? Concretely, please. Jytdog (talk) 23:13, 25 February 2016 (UTC)

Jytdog: the comment wasn't in response to the templating, but to the "random musings of unengaged individuals" horseshit—the "you're not an expert so fuck off" horseshit. If you look closely at the indenting of my comment, you'll see I wasn't intending to respond to you. Curly Turkey 🍁 ¡gobble! 00:31, 26 February 2016 (UTC)

Oh I see; sorry i missed that. elbows do get thrown on this article, more than anyone would like. i'm sorry about that. Jytdog (talk) 00:32, 26 February 2016 (UTC)

Cochrane evidence on cessation should be labeled "moderate"

The article referred to Cochrane, saying "There is tentative evidence that they can help people quit smoking."

However, the evidence that they can help can people quit is moderate according to the cited Cochrane review. In the table titled "Summary of Findings for the Main Comparison", footnotes 3 and 4 clarify that the GRADE rating was downgraded from 'moderate' (relevant to statistical quality) to 'low' for the randomization against placebo, due to the fact that the e-cigarettes in one trial delivered nicotine poorly, meaning the effect size likely underestimates efficacy.

As a binary outcome measure, the evidence that e-cigarettes can help smokers quit is moderate according to Cochrane.— Preceding unsigned comment added by Zvi Zig (talk • contribs) 09:42, 3 March 2016 (UTC)

Source says the confidence level is "low" for a number of reasons. Seems tentative to me. Yobol (talk) 13:18, 3 March 2016 (UTC)

Yobol, you apparently did not read what I wrote above. The GRADE rating had been downgraded from 'moderate' to 'low' because the device in one trial delivered nicotine poorly compared to devices currently on the market, meaning the effect size may be underestimated.Zvi Zig (talk)

Right, the authors say that their confidence is "low" in the effect, which is a another way of saying "tentative". I have no idea where, from the source, you are getting the word "moderate". Yobol (talk) 14:09, 3 March 2016 (UTC)

The word "moderate" is in the table "Summary of Findings for the Main Comparison". It describes the systematic approach to grading evidence.

Footnotes 3 and 4 in the table relate to the randomization of EC against placebo, they read:

• Downgraded one level due to indirectness. The electronic cigarette used in Bullen 2013 was not very effective at delivering nicotine
• Downgraded one level due to imprecision. Only two included studies, small number of events (<300) in each arm Zvi Zig (talk • contribs)

That table describes all of the studies as being of "low" or "very low" quality. I do not see any evidence in the source to support the change you propose to "moderate". Yobol (talk) 15:48, 3 March 2016 (UTC)

Zvi Zig it seems that the argument you are making, is a) from the footnote, it seems that the efficacy was underestimated because the device delivered nicotine poorly b) so the evidence is stronger than it appears; c) so we should say it "moderate" not "low" - is that an accurate description of your argument? Jytdog (talk) 16:22, 3 March 2016 (UTC)

Thanks, Jytdog. Let me clarify. The footnotes quoted above are Cochrane's explanations for the assigned GRADE rating.

On statistical quality, the results are 'moderate'. However, because of poor nicotine delivery, the results (RR 2.29) may have been underestimated and thus the GRADE rating had been downgraded to 'low'.

So, if the question is what's our confidence in the RR, then the correct answer is 'low'. However, if the question is what's our confidence that e-cigarettes can help cessation, then the correct answer is 'moderate'. Zvi Zig (talk • contribs)

Except nowhere in the actual source does it say that the evidence for smoking cessation is "moderate", so we can't say that it is. Yobol (talk) 21:46, 3 March 2016 (UTC)

The systematic rating system is fairly explicit. At the very least, there's no justification for describing the evidence that e-cigarettes help cessation as tentative. Zvi Zig (talk • contribs)

The source explicitly states that their confidence is low as to the effect, and the source does not explicitly state anywhere that the evidence is "moderate". Time to drop the WP:STICK, your version is not supported by the source. Yobol (talk) 22:46, 3 March 2016 (UTC)

To reiterate:

1. The source is explicit that rating was downgraded from 'moderate' due to poor nicotine delivery (in Table titled, "Summary of Findings for the Main Comparison").
2. Regardless of whether the word 'moderate' is considered explicit, or not, the 'low' rating reflects the finding that the effect size may have been underestimated.
3. Being that the the 'low' rating actually reflects the possible underestimation of the effect size, it is inappropriate to use it to support describing the evidence for efficacy as "tentative". Zvi Zig (talk • contribs)

Those points just prove it further – tentative is pretty much the only accurate summation. CFCF 💌 📧 23:21, 3 March 2016 (UTC)

CFCF, you are apparently not differentiating between the effect size, to which the risk of underestimation relates, and the question of whether e-cigarettes are effective. Zvi Zig (talk • contribs)

Thanks for confirming above. I see what you mean, but this is going beyond summarizing what the source says and analyzing it, and coming up with your own new finding. This goes beyond summarizing the source, and is the kind of thing that WP:OR disallows. I would say the same thing if you were downgrading the conclusion. Jytdog (talk) 18:50, 4 March 2016 (UTC)

Thanks, Jytdog. Let me clarify please...

Cochrane's 'low' score is the basis of Wikipedia's assertion, "There is tentative evidence that they can help people quit smoking". However, the 'low' grade was assigned to the effect estimate, not to the (binary) positive outcome. Cochrane specifically states, "These GRADE ratings reflect low levels of confidence in the effect estimates".

I think that the other editors here want to extend the 'low' confidence grade from the effect estimate to the binary positive outcome. This is absurd; the 'low' score stems from the suspicion that the effect estimate is underestimated. Zvi Zig (talk • contribs)

Stating the journal statements are from

Have reverted this edit [3] which added "according to the American Journal of Respiratory and Critical Care Medicine in 2014"

If we begin stating which journal and which year all our content comes from this page will turn into a disaster. Adding the quote from the paper for the second bit. Doc James (talk · contribs · email) 04:47, 6 March 2016 (UTC)

Fair enough. I'm willing to compromise on that one, but as I stated in the edit summary, the year is also needed to give the statement a point of reference. Instead of ignoring my comments and simply reverting (as you did twice), it would be more helpful to drop me a line on my talk page or discuss it here first.

That wasn't the only edit you reverted, by the way. On two occasions, you also restored the statement, "Trying e-cigarettes was common among less educated people.", and this last time you changed "common" to "more common" including the quote from the source in an attempt to address my concern. However, I still disagree that this provides any real value to the article. First of all, the source does not state it is "more common" in the "less educated". It simply points out that it is common, and that's it. Without providing a direct comparison to the "more educated", the statement falls flat of providing any real value to the article and context it was provided in. I propose we remove it. --GoneIn60 (talk) 16:37, 6 March 2016 (UTC)

• Agree, it doesn't provide any real value. The Use section currently repeats many factoids originating from small surveys. With often geographically limited and frequently outdated results, they have been used, either by review authors or Wikipedia editors, to make broad conclusions about ecig users that they don't necessarily support. In this example, the source cited a 2010 US survey of 249 ever-users. Another source used elsewhere in the article said, "Compared with nonsmokers, EC users tend to be younger, better educated, and of higher income class" (from another limited survey). P Walford (talk) 21:03, 6 March 2016 (UTC)

@Doc James: The case to remove some of the content you restored has been made. Care to comment at this point? --GoneIn60 (talk) 10:45, 10 March 2016 (UTC)

With respect to adding the year. I am fine with that. Doc James (talk · contribs · email) 17:36, 10 March 2016 (UTC)

The primary issue deals with the removal of the statement quoted above. If you still oppose its removal, please explain your position in detail, taking into account the arguments supporting its removal. --GoneIn60 (talk) 00:50, 11 March 2016 (UTC)

C&EN Cover Story

The C&EN Cover Story Boom In E-Cigarettes Sparks Calls For Regulation might useful for updating the article. --Leyo 11:15, 7 March 2016 (UTC)

Lead

Doc James -- I think you're making a mild error of judgment on this article. You keep removing paragraph breaks in the lead, and I don't think that's well advised. It's true that WP:LEAD says there should be four paragraphs in the lead. This is not the same as four blocks of text, and what you're actually making is blocks of text that consist of several different paragraphs all jammed together.

A paragraph is a thematically related series of ideas which develop a particular subject. A block of text is the area between two line breaks. Do you see the difference?—S Marshall T/C 12:42, 10 March 2016 (UTC)

Social and cultural aspects are one theme. Health and safety is another theme. The paragraphs are broad themes. Doc James (talk · contribs · email) 17:39, 10 March 2016 (UTC)

Rightly or wrongly, on WP paragraph = block of text. Johnbod (talk) 17:52, 10 March 2016 (UTC)

• So just to confirm ---- the fourth block of text, which begins with the history and geography, goes on to discuss user numbers in Western democracies, and then takes a little side-trek into reasons for use and dual use with tobacco cigs before talking about law and regulation, and then finishes off with brand numbers ---- you don't feel that's incoherent and confusing at all?—S Marshall T/C 19:10, 10 March 2016 (UTC)

Outdated reference inapropriately restored

I had deleted the statement, "There is little data about their health effects", which references a paper authored mid-2013. Using statements about research quantity from a time when the overwhelming majority of today's research was unavailable is inappropriate.

A Pubmed search on e-cig terms shows that only a fraction of today's research had been available at the time.

My deletion had been reverted [[4]] by Yobol who explained, while there is clearly more data, there is still little high quality safety data regarding e-cigarettes, especially long term.

This does not justify restoring the irrelevant source; whatever one's views on today's data quality, this certainly not supported by a mid-2013 estimate on data quantity.

References

- Zvi Zig 11:34, 11 March 2016 (UTC)

Positions of health authorities

I have reverted this set of changes, as they distort the position of regulatory authorities. If you read the WHO statement, it is pretty hostile to current e-cigs - the known risks, the unknown risks, especially the lack of regulation. yes there is the one line that says "the use of appropriately regulated ENDS may have a role to play in supporting attempts to quit." but even there, it is not current ENDS, but rather "appropriately regulated" ENDS, which do not exist, as the report makes clear. So I reverted. Happy to discuss. Jytdog (talk) 14:43, 11 March 2016 (UTC)

Jytdog, you are right with regards to the WHO report. This was an embarrassing terrible mistake on my part.Zvi Zig (talk • contribs 11:08, 13 March 2016 (UTC)

:). no problem, thanks for being gracious. Jytdog (talk) 18:31, 13 March 2016 (UTC)

Section ordering

Not sure what TracyMcClark is talking about here with regard to "Blind revert"? The ordering of this article has been talked to death:

• 2015-05 Talk:Electronic_cigarette/Archive_24#Section_ordering (which has review of the extensive past discussions and links to them.

P Walford's change in this dif was too bold per the DS as I noted when I reverted in this dif. This is not something people should just be messing with, without prior discussion. Jytdog (talk) 20:38, 13 March 2016 (UTC)

The order of the sections has been an ongoing issue and the order has been stable for a long time, the result of a consensus discussion, carried out by an uninvolved admin. If the sections are reordered it should be the result of a discussion/RFC. Be aware that the last two or so RFC's ended as no consensus. This probably has more to do with the nature of the device. It is not a medical device and making the page in a medical order or something close is not likely to reach consensus. AlbinoFerret 20:41, 13 March 2016 (UTC)

(ec) - Jytdog: Have you again not paid attention? The longstanding order was first changed yesterday w/o any further discussion.--TMCk (talk) 20:45, 13 March 2016 (UTC)

Agreed P Walford's edit simply restored the long standing page order. AlbinoFerret 20:47, 13 March 2016 (UTC)