Alexuang

Hello, Alexuang, and welcome to Wikipedia! My name is Shalor and I work with the Wiki Education Foundation; I help support students who are editing as part of a class assignment.

I hope you enjoy editing here. If you haven't already done so, please check out the student training library, which introduces you to editing and Wikipedia's core principles. You may also want to check out the Teahouse, a community of Wikipedia editors dedicated to helping new users. Below are some resources to help you get started editing.

If you have any questions, please don't hesitate to contact me on my talk page. Shalor (Wiki Ed) (talk) 18:56, 29 July 2019 (UTC)Reply

Hi! I saw that you used some studies in the Kwashiorkor article. This poses a bit of an issue since they weren't accompanied by secondary sources that backs up the claims.

The issue with studies is that they're primary sources for the data compiled by its creators, who are also typically the ones who conducted the research. The publisher can't serve as a verification of the study, as they only review to ensure that there are no obvious errors that would invalidate the study and determine if the study is something that should be published. They don't provide any commentary, which a secondary source would. Studies are also fairly limited out of necessity, as there's almost always not possible to survey every person that could fall into the study area. As such, the findings are seen as only really true for those persons who participated. The results could differ if they were to survey people from a different area, with a different ailment than cancer, or from different backgrounds or income levels. The secondary source will help provide that context. You could also have someone ask why one study was chosen over another that may give different results or one that isn't in English. That said, it is generally OK to use the literature reviews in a study as long as you're careful to avoid the parts where the author uses them to discuss their study.

Essentially what you need to do here is to add secondary sources to the article. This is all especially important when it comes to discussing health related data and claims. The policy page on the guidelines for sourcing medical articles goes over this in a bit more depth.

I hope this helps! Shalor (Wiki Ed) (talk) 19:19, 5 August 2019 (UTC)Reply

Hi Shalor,

I briefly reviewed my citations and the most problematic ones I identified were the Malawian twin study (Smith et al 2013) and the Pham et al 2019 paper. These are both already supported by the paper by Briend (2014) which, in my reading, is a reliable secondary source. I have also just added in another review article on the gut microbiome by Velly et al (2017) that touches on the 2013 Smith et al paper in the broader context of our current understanding of microbiome changes and dysbiosis.

Can you clarify what you mean by, "That said, it is generally OK to use the literature reviews in a study as long as you're careful to avoid the parts where the author uses them to discuss their study."? I'm not sure I understand this sentence.

On a related note, I understand that we should not rely on primary sources for wiki health articles but I'm not sure where the distinction lies between being able to properly cite/draw information from a review article discussing the primary literature vs. not being able to directly cite from a primary source. Since review articles commonly paraphrase the primary literature they are reviewing, are we only allowed to draw from the commentary/conclusions of the reviewer? If I were to utilize a primary source alone as a reference for a certain statement (which I understand I should not do), does the nature/context of that statement matter? I can understand why it would not be a good idea to present information as a claim if it is the conclusion of a research study, but what if information is simply being stated in a more neutral manner?

For example, for the kwashiorkor article I wrote "Several hypotheses have been proposed that are associated with and explain some, but not all aspects of the pathophysiology of kwashiorkor. They include, but are not limited to protein deficiency causing hypoalbuminemia, amino acid deficiency, oxidative stress, and gut microbiome changes." To me, this is a neutral statement that is different from, say, a hypothetical claim that "a 2013 study in elderly males showed that eating fish twice a day lowers the rate of heart attack by 23%". I can understand why having a health claim on a health article based solely on one primary source would be a bad idea, but does this same standard apply to more 'matter-of-fact' information that does not make the same kind of health claim?

Thanks for any insight you can provide.

Cheers, --Alexuang (talk) 23:38, 6 August 2019 (UTC)Reply

• Hi! Sorry for replying late - the best way to make sure I see a reply is to tag me using this: {{u|Shalor (Wiki Ed)}} on the talk page.

With the literature reviews in a study, what I meant with this is that you can use basic statements that go over the literature, however we cannot use statements where they tie that other literature into their study. Here's an example:

A researcher is conducting a study on the impact of vitamin C on the common cold. Their goal is to show that taking vitamin C each day will minimize chances of getting sick and the amount of time having the cold. They review other literature, such as studies that also look at how vitamins impacts health. One of these is a study that showed that taking vitamins along with a balanced diet resulted in better overall health in the study participants. Now when it comes to the researcher's review of the topic, we can use things such as "The results showed that the combination of vitamins and balanced diet prompted an increase in participant health, which correlates with Jones's 2015 study on the same topic." This is something that is stating a fact from the study and cross references it with another study that was on the same topic area. However what we can't use are statements along these lines: "The study showed that the use of vitamins and a balanced diet improved participant health, which reflects on our hypothesis that a daily dose of vitamin C in specific will improve health by doing this and that."

Basically, it's when they start moving from summarizing or contrasting the literature against each other to comparing and contrasting it to their own study that it becomes something to avoid unless you can find a secondary source.

Now when it comes to the second part of the question, I'm going to confer with one of my colleagues who is more familiar with this subject area and get back to you on this. Shalor (Wiki Ed) (talk) 17:42, 15 August 2019 (UTC)Reply

• I actually went ahead and posted to the medicine WikiProject to get their call on this as well. I think this is one of the grey areas of medical sourcing guidelines, so it would be a good idea to get their perspectives as well. Shalor (Wiki Ed) (talk) 17:53, 15 August 2019 (UTC)Reply
  • This looks like a question about using a primary source (e.g., comparing twins to each other and writing about what you see) that contains a "Background" or "Introduction" section. That section is technically a secondary source ...but in practice, I recommend against using it if possible.
    Alexuang, I noticed that several of the statements in the Kwashiorkor article have a lot of citations after them. One is usually enough, and more than three is usually a sign of problems. So one thing you could do is to look at the sources that you're uncertain about, and see whether they need to be used at all. For example, the twins study is a primary source (overall), so if the twins study is the third (or fifth) citation after a sentence, then maybe just take it out. Or if it is the only citation, but there is a recent review article or recent textbook that says the same thing, then try replacing it with the review article. Once you've got it narrowed down to any "necessary" cases, then we can talk about whether that specific source is ultimately reliable for the specific statement. But first I think I'd minimize the number of times you have to deal with that evaluation process. :-) WhatamIdoing (talk) 21:52, 15 August 2019 (UTC)Reply