Wikipedia:Articles for deletion/Smoker's paradox


 * The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review).  No further edits should be made to this page.

The result was merge to Health effects of tobacco. Selectively and carefully, as discussed below.  Sandstein  09:07, 18 April 2021 (UTC)

Smoker's paradox

 * – ( View AfD View log )

After reviewing the article's citations and searching for secondary sources on PubMed, one draws the conclusion that Smoker's paradox (SP) is not a term used in the literature for all cases of a possible inverse correlation between smoking and a disease's incidence or severity; it is rather almost exclusively used for cardiovascular diseases, and COVID19 as of late. There is only one source, and actually far from what is considered a good medical one, that uses the term for general in-hospital mortality. (In PubMed there's a study of trauma mortality drawing the conclusion of a "SP"; and there are some articles referring to the "smoking paradox" in the blunted correlation between smoking and lung cancer in Asian populations, and in psoriasis. ) So the definition of SP in the article is unsubstantiated, and the article's current area of focus is wider than can be attributed to sources. Any synthesis of information from existing sources to talk about a wide ranging SP concept - like the current article does - would be a type of original research.

In the groups of diseases where the existence of SP has been hypothesized, namely cardiovascular diseases and COVID-19, it is far from certain. As a statistical phenomenon, SP reappears in medical literature in the last 25 years for various findings, which are for the most part either attributed to confounding factors or eventually refuted. Furthermore, it has currently no clinical significance, as nobody has ever suggested that patients should not quit smoking, and no treatment implications have been accepted. Even so, some very careful mention of SP as an unconfirmed concept could be considered in health effects of tobacco, myocardial infarction or myocardial revascularization, but the body of literature definitely doesn't warrant an independent article. Given that the majority of sources are primary ones, merging of the article as it is to health effects of tobacco doesn't seem appealing either.

I have notified the article's creator of the issues, and they kindly recognized them. NikosGouliaros (talk) 14:49, 3 April 2021 (UTC)
 * Note: This discussion has been included in the list of Medicine-related deletion discussions. Shellwood (talk) 14:54, 3 April 2021 (UTC)
 * Note: This discussion has been included in the list of Health and fitness-related deletion discussions.  Spiderone (Talk to Spider) 19:04, 3 April 2021 (UTC)

 Relisted to generate a more thorough discussion and clearer consensus.

Please add new comments below this notice. Thanks, Eddie891 Talk Work 23:32, 10 April 2021 (UTC)


 * Merge to health effects of tobacco (only), and then remove nearly all of the contents so that it amounts to "some very careful mention of SP as an unconfirmed concept". The ==History== section might be the best location. WhatamIdoing (talk) 02:40, 11 April 2021 (UTC)
 * Merge per WAID. Vaticidalprophet 07:51, 11 April 2021 (UTC)
 * Merge per WAID. 31.50.193.212 (talk) 11:29, 11 April 2021 (UTC)
 * Adding In trying to clarify the content in line with a more historical perspective, the page has grown a bit. Since I'm epidemiologically nerdy enough to find this sort of thing informative - and in the case of a Parkinson's disease genuinely intriguing - I'm ultimately agnostic about whether it's better to MERGE or KEEP. (Fwiw, if others decided for KEEP, I could try to improve the page; on the other hand I wouldn't care to have tpo do the merging :-) 31.50.193.212 (talk) 15:17, 11 April 2021 (UTC)
 * I feel the need to repeat that the term "smoker's paradox" has never, to my knowing, been used in Parkinson's disease (PD), or indeed anywhere else on a repeated, long-term basis, outside cardiovascular disease. Given that smoking has not been implicated in the pathophysiology of PD, the reverse correlation found is not a true paradox. For this reason, the article in its current state or in the state it was when proposed for deletion has almost nothing to merit adding to health effects of tobacco. "Smoker's paradox" has not been used in reliable medical literature as an umbrella term to refer generally to the finding of inverse correlation between smoking and any given disease; this means that even after all the work IP and other contributors have put to it, it remains heavily laden with primary synthesis of published material. Merging it is unwarranted. NikosGouliaros (talk) 19:16, 11 April 2021 (UTC)
 * NikosGouliaros, my initial reaction is that I quite strongly disagree with your claims as representing (it would seem, no offence :) your own (unsupported) pov. However, having worked on a page that - coming as I do from a background of medical writing in the field of epidemiology - appears to me intrinsically interesting and possibly educational I feel that I now have some sort of an individual coi, and I wish to recuse myself from the !vote (full disclosure: I was intending to do this anyway, perhaps after sleeping on it). I absolutely understand where people are coming from who may be concerned that the page could be construed as a medical apology for smoking, but I don't think any gf reader could honestly claim that in its current form. Ultimately, I suspect that this sort of decision may come down to individual exclusionist/inclusionist tendencies; fwiw, my own rule thumb goes something like "if, as a lay reader, I was looking for reliable information on this subject, would I find the present page helpful?" And here I feel the answer could be "yes", which is why I've tried to improve the page. Perhaps I should add that the issue was by no means new to me: for decades, when people asked me questions like "isn't smoking beneficial for anything healthwise?" I've answered "just possiblyif you know that you're at high risk of developing Parkinson's disease, a little..." Cheers, 31.50.193.212 (talk) 21:02, 11 April 2021 (UTC)
 * No offense taken. I only cannot see what point of view I can be thought of representing, other than that the article, in its current form (let alone as it was before much of material from primary sources was removed), is unsupported by reliable sources with regard to the SP. On the other hand, I see how my writing that the article has nothing worth merging could be misunderstood: I meant nothing worth merging under the title "Smoker's paradox". I clarify that I consider the included information on Parkinson's perfectly worthy of inclusion in Wikipedia (and there's already a mention in Health effects of tobacco), just not in connection to the SP concept; this and only this is what I consider synthesis of published material: describing the effects of smoking on PD under the title "SP". Again, SP, if we stick to reliable sources, can't be the title of any text on the positive effects of smoking in general. Finally, I note that I have expressed worries I have never expressed worries that the article may become an apology for smoking; and I cannot see any COI in your vote. NikosGouliaros (talk) 21:43, 11 April 2021 (UTC)
 * Thank you for your thoughtfulness and kind words Nikos. Always appreciated. 86.186.168.248 (talk) 21:42, 12 April 2021 (UTC)


 * Keep (but maybe rename). It might be a dubious concept, with only scant mention in literature and maybe even only older lit if it's been discarded as a theory or no longer of interest, and the term itself might not be specifically used for some effects that are now discussed in the article. But either taken as the strict literature use, or as the broader sense, it's a well-defined topic that does have mention in the literature and seems quite different than the concepts covered in the main health-effects article. And it's a novel concept, that we can write about dispassionately. I fundamentally reject that merely having this as its own article is any sort of advocacy or appologetics; WP:N is how we know. Therefore, I think it should be a self-contained sub-article as long as we have sources that support it. I agree we need reliable sources, but I don't think the past-decade limit of MEDRS is appropriate if the whole concept was only studied prior to that timeframe. It's an editorial decision whether "smoker's paradox" should be kept as the strict set of cases where this term is applied in the literature or as a wiki-editor/lay-language term for the broader scope. And if broader, should we instead use a different lay-language term to avoid mixing it up with the medical term. DMacks (talk) 21:43, 11 April 2021 (UTC)
 * Here's another recent pooled-studies article in the coronary arena: 10.1016/j.jcin.2019.06.034. So we have apparent MEDRS for this aspect and limited meaning of the phrase. Do we hae MEDRS for the COVID aspect? We have at least some sources for use of this term in that context. And we have other sources for use of this term in other-other contexts. So I don't think it's SYNTH to use this phrase broadly. DMacks (talk) 02:00, 12 April 2021 (UTC)
 * This is (important) primary research on SP in cardiovascular disease. NikosGouliaros (talk) 13:08, 12 April 2021 (UTC) NikosGouliaros (talk) 13:08, 12 April 2021 (UTC)


 * Keep, but maybe rename... but I'm not against a sensible merge either. There is a well-referenced history of results around smoking and cardiovascular diseases that tell a complicated but important epidemiological story. As per, this is not apologetics. I agree with others here that we need to be careful about the name. I would remove other potential beneficial results of smoking, as with Parkinson's, from here and expand Health_effects_of_tobacco. Health_effects_of_tobacco could also say something about the value of nicotine in the treatment of colitis, for example. Bondegezou (talk) 08:59, 12 April 2021 (UTC)
 * Bondegezou, I agree with your point about the epidemiological story (assuming Wikipedia has the ability, going forward, to curate/update per WP:MEDDATE). But why exclude Parkinson's disease if this is also well-referenced and a continuing area of research. 86.186.168.248 (talk) 11:50, 12 April 2021 (UTC)
 * Because it's not part of the same story. We have MEDRS-compliant citations to support the epidemiological story of the "smoker's paradox" about cardiovascular disease and smoking. We also have MEDRS-compliant citations about the potential beneficial effects of smoking/nicotine in Parkinson's and in colitis, but those are not called the "smoker's paradox", reflect different phenomena and are not linked together by MEDRS-compliant citations. Bondegezou (talk) 12:19, 12 April 2021 (UTC)
 * Bondegezou, Ok I can see where you're coming from now (and that consideration seems to feed in to some discussion here about the appropriateness, or otherwise, of the page name). Searching PubMed for "smoker's paradox" or "smoking paradox" without  filtering for reviews etc (in the belief WP:MEDRS may not apply to this basically non-biomedical, lexical question - ping WAID for her opinion) indicates that although usage of the term(s) appears most common for cardiovascular disease, that isn't entirely the case (and, for example... would paradoxes regarding ethnicity fit the scope of the page? cf 28092929, etc). Fwiw, my development of the PD question within the apparent scope of the page (WP:OOP?) was based on 1) the broader-brush definition in the lead, which, however, currently remains unsourced; 2) similarities in the underlying methodological issues regarding the interpretative challenges posed by a series of apparently consisistent but paradoxical findings derived from observational research on smoking (even when the term paradox etc isn't explicitly used); and 3) perceived topicality. Clearly, I'm too involved by now with the Parkinson's content to be independent here, haha. 86.186.168.248 (talk) 15:03, 12 April 2021 (UTC)


 * Merge to health effects of tobacco. All the content can be covered there. --Spyder212 (talk) 21:26, 12 April 2021 (UTC)
 * Merge - I agree that medical sources aren't necessary to support vocabulary; the term can be mentioned in passing. Any positive correlations should clearly be mentioned in the health effects article anyway. -- Beland (talk) 19:10, 17 April 2021 (UTC)


 * The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.