Wikidata:Property proposal/may prevent
may prevent edit
Originally proposed at Wikidata:Property proposal/Natural science
Description | disease which may be prevented by this drug |
---|---|
Data type | Item |
Domain | Medicine |
Allowed values | Wikidata items of instance of (P31) or subclass of (P279) disease (Q12136) |
Example | aspirin (Q18216) → myocardial infarction (Q12152) --> |
Source | http://www.fda.gov/, http://ema.europa.eu/, https://www.nlm.nih.gov/research/umls/sourcereleasedocs/current/NDFRT/ |
Planned use | several hundred preventative drug usages should be added to Wikidata |
See also | drug or therapy used for treatment (P2176), medical condition treated (P2175), has effect (P1542), significant drug interaction (P769) |
- Motivation
As listed above, Wikdata has several properties to describe drug-disease interactions (drug or therapy used for treatment (P2176), medical condition treated (P2175)), drug side effects (has effect (P1542)) and drug-drug interactions (significant drug interaction (P769)). What is still missing is a way to describe that a certain medication can prevent a disease from even occuring. This property proposal should cover these cases. Sebotic (talk) 19:00, 8 February 2017 (UTC)
Notified participants of WikiProject Medicine
- Discussion
- What exactly is meant by "may"? I think the property should likely to be worded to be more clear about the meaning. ChristianKl (talk) 22:09, 8 February 2017 (UTC)
- I followed the wording of NDF-RT and it makes sense, because medications in many cases do not have a 100% success rate. But certainly, I can rename it to 'used for prevention of', if you prefer that. Sebotic (talk) 09:54, 9 February 2017 (UTC)
- Many substances that are labeled with drug or therapy used for treatment (P2176) aren't better than placebo's. They are used by doctors to treat but there's no evidence that they are effective for that purpose. When it comes to this property it's important to be clear about whether this property indicates that there's evidence that supports this claim. I don't think it has to be made clear in the property name so "may prevent" is okay. But it should be made clear in the property description. We don't want people to make bad decisions because they think the property indicates proof for the capability of preventing a disease when it doesn't. ChristianKl (talk) 14:50, 10 February 2017 (UTC)
- support the creation of this property and feel it is needed, but prefer the 'used for prevention of' label to be consistent with the way the treatment property is written. Gtsulab (talk) 18:01, 1 June 2017 (UTC)
- I see this is medicine-specific variant of "used to prevent"
- Support both of them (specific to medicine and other) because medical relations are for specific situations (very rare on occasion) d1g (talk) 03:28, 15 August 2017 (UTC)
- Support Should not be limited to just drugs. May means it does not always prevent. Doc James (talk · contribs · email) (if I write on your page reply on mine) 21:14, 19 September 2017 (UTC)
- @Doc James: To me that clarification doesn't help much. If we have a clinical trial that shows that consuming lettuce is correlated with a lesser risk of heart disease, is that sufficient evidence for this property to be used to store the resulting information? ChristianKl (talk) 11:28, 12 October 2017 (UTC)
- This sounds more like and more like an issue with data sources as opposed to an issue with the property itself. An FDA approved indication seems like a more reliable source than a clinical trial, and the issues you've raised about the 'may prevent' property apply to the 'drug used for treatment of' property (which has already been approved). Would it help if this property were reworded to 'used for the prevention of' and constrained to just drugs? But the FDA does investigate medical devices and procedures too. Gtsulab (talk) 16:18, 19 October 2017 (UTC)
- @Gtsulab: 'drug used for treatment of' has been approved but we had drama with it that resulted in the property getting removed from enwiki infoboxes for including data on drugs with where in clinical trial stage 1/2/3. This episode damaged the reputation of Wikidata in the EnWiki Project Medicine. It's drama I wouldn't want to repeat and the current formulation of this property appears to me to have the danger of producing problems. ChristianKl (talk) 09:54, 29 October 2017 (UTC)
- This sounds more like and more like an issue with data sources as opposed to an issue with the property itself. An FDA approved indication seems like a more reliable source than a clinical trial, and the issues you've raised about the 'may prevent' property apply to the 'drug used for treatment of' property (which has already been approved). Would it help if this property were reworded to 'used for the prevention of' and constrained to just drugs? But the FDA does investigate medical devices and procedures too. Gtsulab (talk) 16:18, 19 October 2017 (UTC)
- Support --Okkn (talk) 21:00, 30 December 2017 (UTC)
@Sebotic, ChristianKl, Gtsulab, D1gggg, Doc James, Okkn: Done Micru (talk) 07:49, 15 March 2018 (UTC)