Misplaced Pages

Revision as of 22:41, 6 June 2009

Benzodiazepine

Nominator(s): Literaturegeek | T@1k? 08:31, 26 May 2009 (UTC)

• Featured article candidates/Benzodiazepine/archive1

I am nominating this for featured article because I believe that the article has finally reached the criteria for featured article. It is indepth and comprehensive, makes extensive use of high quality recent sources, written clearly and consisely. I also believe that the article has the right balance where both professional and layman can derive a great deal of knowledge and understanding from the article subject. Of course there may be some suggestions made before it is promoted to featured article and I am open to constructive criticism or suggestions for improvements but I do believe the article is ready for featured article nomination for review and hopefully promotion.Literaturegeek | T@1k? 08:31, 26 May 2009 (UTC)

Restart, old nom; images and reliability of sources checked. SandyGeorgia (Talk) 19:19, 6 June 2009 (UTC)

• Oppose. The sourcing is not to a professional standard: too many citations per fact; too many old papers cited; non-relevant-focus papers cited; inconsistency between sources used for the lead and the body; primary research papers cited without good reason; sources of narrow scope used where wider scope required. The article is apparently comprehensive (I'm no expert) but not discriminatory enough in the facts it chooses to present. The prose isn't consistently at FA standard and needs a copy edit (but don't advise one at this stage due to other problems). Two prose examples: there are a number of single-sentence paragraphs and paragraphs beginning with "They". The "Mechanism of action" section is utterly impenetrable to anyone without a pharmacy degree, which isn't acceptable. We must attempt to explain to the general reader how benzos work in the brain and achieve the effects they do. Why do they make you less anxious? Why do they make you sleepy? Why do they relax muscles? Why do they stop seizures?

Since sources are the foundation upon which an article is built, I don't believe these problems can be fixed quickly. The article shows some evidence of having random fact laid upon fact rather than being planned, executed and refined. The editors need to go through each section, selecting a small set of high quality and focussed sources, and build the text from what those sources say and the weight the sources afford to the facts. As an example of this practice, the seizure section was reduced from thirteen sources to three. I don't suggest this FAC be withdrawn despite my opinion that it can't be rescued in time: I hope we can get reviews by editors with expertise in English and in Medicine. Here are some (but by no means exhaustive) examples of specific problems found:

1. The first three sentences of the third paragraph in the lead are cited to three separate papers. But the text is standard stuff that surely could be cited to one review? All three of those are written by Lader and one paper is ten years old.
2. The next sentence on long term use cites three sources, two of which are old (20 and 18 years) and are focussed on withdrawal rather than long term use.
3. The two sentences on pregnancy in the lead are sourced completely differently to those in the body text. The lead uses sources that are 11 and 15 years old.
4. The lead text on the elderly and on drug overdose is cited to different sources than the body text.
5. The statement on the general properties of benzodiazepam is cited to a primary research paper on genetically modified mice.
6. The statement in "Therapeutic uses" on the various administration routes is sourced to a primary research paper from 1978 on five newborn infants with convulsions.
7. Source PMID 7388368 is just the sort of paper we should be citing, except it is 30 years old. Surely there is a more up-to-date version?
8. The lead statement on their properties and uses in veterinary practice is sourced to two papers: A review from 1975 on restraint methods in swine (which mentions diazepam for minor procedures) and a Bulgarian-language research paper from 1984 on the effects of powdered diazepam on various animals. Surely a vet textbook could be used, and be more clinically relevant and up-to-date. BTW: there doesn't seem to be any justification for citing non-English sources for this article.
9. The statement "Midazolam can also be used along with other drugs in the sedation and capture of wild animals." is sourced to a primary research paper that captured nine aardvarks with ketamine combined with one of five other test drugs, one of which was midazolam.
10. The (vet) statment "They are used before surgery as premedication for muscular relaxation" is sourced to a 20 year old primary research paper on five dogs undergoing anaesthesia. This doesn't tell us about modern vet practice. Indeed, none of the sources in the vet section could be used to document how benzos are currently used in vet practice.
11. The statement "Benzodiazepines are also used to treat the acute panic caused by hallucinogen intoxication" is sourced to a 20-year old paper on hallucinogenic intoxication, the abstract of which says "Panic reactions may require treatment with a benzodiazepine or haloperidol." Benzos have well-known anti-anxiety properties so of course they will be used for severe panic attacks. Is this really notable, or something random a PubMed search found?
12. The drug related crime section contains a long paragraph sourced to a single primary research study on Australian police detainees. My gut feeling is that the some of the results here may be particular to Australia during the period studied, and give different results for other countries or different demographics.
13. The statement "Eclampsia also responds to them but benzodiazepines are not as effective as intravenous magnesium." is included in the "Other indications" section, but is sourced to a review of diazepam (only, so not "them") vs magnesium. The use of diazepam here is no different to its use in treating any prolonged seizure, so this isn't an "other indications". In fact, much of the "other indications" deals with anxiety situations, and we have an axiety section. And where's the muscle-relaxant indications (only tetanus and stiff person syndrome are mentioned, but "muscle spasm of varied aetiology" are an indication for diazepam)?
14. There are far too many cases where three, four and even five citations are strung together, sometimes just to source a single sentence. I'm not sure what the cause of this is, but it isn't necessary for a subject like this. We are dealing with an mature class of drugs for which there is an abundance of high quality literature. Occasionally, a sentence or paragraph may require multiple sources when combining distinct facts, but that doesn't seem to be the case here. I don't know whether the editors are trying to strengthen an argument by citing multiple sources but that isn't necessary on WP. The other possibility, and more worrying, is that multiple sources are being combined to produce a statement that may transgress WP:OR. I'm particularly concerned about the various places side-effects/symptoms are mentioned. It is not the job of Wikipedians to pool the side effects (or withdrawal symptoms) of the various benzos and come up with a list of notable ones.

(BTW: Please can you respond below my signature rather than within the above text, to avoid breaking it up. Thanks.) Colin° 21:23, 6 June 2009 (UTC)

Thank you for taking the time to review the article and making these recommendations.

Point 1. I did actually try to resolve that by locating review papers on pubmed. Actually an idea has popped into my head, I have a text book on benzodiazepines, a reliable one. It is written by Roche pharmaceuticals (I think). The entire book is about benzodiazepines, it is bound to cover these issues in the first paragraph, I just hope that I can locate them easily without having to read the whole book LOL. I shall see what I can do.

Point 2. I can get more recent citations for long term effects but unfortunately they would be primary sources I think. Very little research money is given to study long term effects of benzos, much of the research was carried out back in the 90's. I would be of the opinion that the citations should stay if they cannot be replaced with more recent secondary sources. Perhaps we could remove it from the lead though and just discuss it briefly in the article body somewhere?

Point 3. I believe that this can be resolved with a bit of effort.

Point 4. See point 1 for resolution.

Point 5. See point 1 for a resolution.

Point 6. I am not aware of any update to the guidance of 1980 by the MHRA or related bodies, well there was actually a small update in 1988. The problem is these regulatory bodies only do updates if they believe the evidence base has changed or their views have changed and need to be revised. If they don't think policy needs changing they just leave things as they were. How do we resolve things like if current national guidelines are based on old reviews? I say they are current because the BNF, Department of Health still use these guidelines so they are current in that they are still functional review and guidance but only because it hasn't been updated. I think that we should stick with that until it is updated by the MHRA.

Point 7. I know very little about veterinary medicine and am not much more than a layman. I also do not have access to any vet books. Is there a wiki vet project that we could contact? Otherwise I think that improving these issues is not going to be very fruitful. perhaps we should delete the vet section. I dunno if I want to do that but if it is not possible to resolve it and the section is holding the FA back, oh well maybe it will have to be deleted.

Point 8. Done. Deleted the primary source.

Point 9. Hmmm, I am beginning to think that we either need to contact a wiki vet project (if it exists) or else delete section?

Point 10. Done. I deleted the ref.

Point 11. The problem with citing crime data is that these types of data are compiled usually by or on behalf of governments rather than clinical researchers so review papers are hard to track down.

Point 10. Partially resolved. Deleted the ref. Other point regarding use as muscle spasm of various causes can be treated with diazepam is a good point. Should we use the BNF as a citation for this?

Point 11. I do have to admit that I sometimes add 2 or 3 sources together. The reason is because is sometimes people challenge single refs as being "only the opinion" of one book, one author or group of authors etc. So sometimes I think that it is necessary if covering an area not widely known or even perhaps controversial or as noted earlier just to make the sentence or paragraph more comprehensive by including additional data, due care though needs to be taken not to do OR of course. Oh and another reason (perhaps bad habit or is it good?) is I tend not to edit articles by putting refs in middle of sentences like after a comma and prefer to add them both to the end of the sentence. Part of the reason the side effects is full of citations is actually because of this review process as there were disputes over sources and which ones were the best etc. Although there were several refs used even before the FA review but they have doubled since it I think.

All valid points Colin. I agree with you that we really don't need to use primary sources for the article but I am not sure if we can achieve 100% no primary sources in the article without article content suffering. There are certain aspects of benzos which are not regularly reviewed or even researched or updated and thus finding recent reviews for certain aspects of benzos is not always possible in my opinion. The crime stats is the first thing that jumps to my mind as something which is rather difficult to find review articles as it is mostly gov reports and stuff and not discussed or reviewed or undergo meta-analysis in the literature for a number of reasons which I can explain if necessary. I do think that primary sources though should be debated and almost always if not always removed if a more review paper exists for that aspect of a subject and if agreed on talk page. I have deleted a couple already that you have suggested and more will probably deleted.

Ok those are my opinions which may be a bit blurred as I am half asleep. I will look forward to other's comments on the FA.--Literaturegeek | T@1k? 22:41, 6 June 2009 (UTC)