Mine PMC for ethics statements

[Daniel-Mietchen]

possible search terms:

• ethical
• "institutional review board"
• "informed consent"
  etc.

[Daniel-Mietchen]

The main purpose here would be to see

• what percentage of articles have a dedicated ethics section, and how that changes over time
• what kind of information is provided in addition to statements of the "... received ethical approval" and "gave informed consent" kinds.
• to what extent PIDs are being used in there and for what, and how that changes over time.

[Daniel-Mietchen]

A simple query for "approval number" currently yields 11404 hits:
https://www.ncbi.nlm.nih.gov/pmc/?term=%22approval+number%22

[Daniel-Mietchen]

Just to clarify that conflict of interest statements are within scope here as well.

[Daniel-Mietchen]

I just reran that "approval number" query from Oct 12, 2017, and it now yields 37963 results, i.e. an about 3.5-fold increase in about 3.5 years.

In the meantime, I have begun to collaborate with @petermr, and we are trying to use his ContentMine pipeline (which is currently being ported to Python) to extract ethics statements from PMC. On the way, we have built a first — still very rough — dictionary (i.e. a set of words highly indicative of the topic of ethics statements), and we are trying to also get a list of ethics committees mentioned in PMC-indexed papers.

[Daniel-Mietchen]

Meeting on April 29, 2021:

• We are considering to submit something to Wikidata Workshop
• We are also considering to submit a Research Idea to RIO and a research paper as well, perhaps in WikiJournal
• There is an event being planned for a weekend in May that is about introducing people to Wikidata in a playful manner. Peter will think about aligning it with the Wikimedia Hackathon
• We also looked a bit into ContentMine dictionaries.

[Daniel-Mietchen]

Some more notes on this by @ShweataNHegde sit at https://github.com/petermr/dictionary/wiki/Ethics-Statement-Project .

[Daniel-Mietchen]

A search for "approval number" now gives 38437 results, i.e. about 500 more than just two weeks ago.

[Daniel-Mietchen]

There are ambiguities at multiple levels.

For instance, this article states that

This study was approved by the Johns Hopkins School of Medicine IRB, Approval Number: IRB00151734. 

The problem here is that Johns Hopkins School of Medicine runs multiple IRBs, and there does not seem to be a straightforward mechanisms to resolve the approval number to get more metadata about the process.

[Daniel-Mietchen]

There is a Office for Human Research Protections (OHRP) Database for Registered IORGs & IRBs, Approved FWAs, and Documents Received in Last 60 Days that has identifiers for IRBs, but these do not resolve either.

[petermr]

I have started to test the phrase extraction tool NLTK-RAKE. https://towardsdatascience.com/extracting-keyphrases-from-text-rake-and-gensim-in-python-eefd0fad582f As with all language tools it will take a day or two to see how useful it is.

…

On Mon, May 10, 2021 at 4:03 PM Daniel Mietchen ***@***.***> wrote: There is a Office for Human Research Protections (OHRP) Database for Registered IORGs & IRBs, Approved FWAs, and Documents Received in Last 60 Days <https://ohrp.cit.nih.gov/search/irbsearch.aspx> that has identifiers for IRBs, but these do not resolve either. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#499 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAFTCSYDBWTBXH7DUKKFQMLTM7YUNANCNFSM4D5M32KA> .

-- Peter Murray-Rust Founder ContentMine.org and Reader Emeritus in Molecular Informatics Dept. Of Chemistry, University of Cambridge, CB2 1EW, UK

[ShweataNHegde]

https://colab.research.google.com/drive/1sFj07mE2XRyeaplvsTs34-VaDHBjnt6U?usp=sharing

Ayush (openVirus volunteers) and I wrote a piece of code that can extract common phrases from a text file with manually scraped Ethics Statements.

[Daniel-Mietchen]

Some updates from this week:

• @ShweataNHegde has created a more refined ethics dictionary here, as per these notes
• I have created Wikidata lexemes for most of the entries in her dictionary, as per this overview
• I also started WikiProject Ethics.

[Daniel-Mietchen]

For more recent updates, see the notes over at Shweata's page.

[Daniel-Mietchen]

Here is a list of ethics-related entities Shweata has mined from articles on stem cells.

[Daniel-Mietchen]

Some more observations by Shweata and Peter sit here.

We now have a dedicated organization, repo and wiki:

• https://github.com/FAIR-ethics
• https://github.com/FAIR-ethics/PMC-ethics
• https://github.com/FAIR-ethics/PMC-ethics/wiki

[Daniel-Mietchen]

The paper How does nursing research differ internationally? A bibliometric analysis of six countries. has a Table 1 that looks at certain features of previous studies, including

Extracted specific properties (e.g., contains ethics statements)

[Daniel-Mietchen]

The project with Shweata and Peter (and Ayush) has since led to a publication:

Hegde SN, Garg A, Murray-Rust P, Mietchen D (2022) Mining the literature for ethics statements: A step towards standardizing research ethics. Research Ideas and Outcomes 8: e94685. https://doi.org/10.3897/rio.8.e94685 .

It outlines a workflow for mining ethics statements and discusses motivations, applications and complications.