Hannah's web log

A recent report from Inside Medicine revealed that CDC scientists received orders to retract or pause publication of journal articles under review or accepted for publication. The directive specifically prohibits use of certain terms: "Gender, transgender, pregnant person, pregnant people, LGBT, transsexual, non-binary, nonbinary, assigned male at birth, assigned female at birth, biologically male, biologically female."

Truly alarming news. But this isn't about my hot take on the horrific censorship going on in the Trump regime. I'm not American and feel a bit out of my depth even commenting. While the political implications are significant, this analysis focuses on a specific technical aspect: the inclusion of "gender" in the prohibited terms list. This common demographic descriptor, often used interchangeably with "sex," appears frequently in scientific literature when describing study populations. Just to prove a point, I set out to demonstrate what a search for these restricted words looks like in PubMed, the largest and most commonly accessed medical database.

A boolean OR search combining all restricted terms yielded 501,185 publications. As anticipated, "gender" dominated these results with 477,056 occurrences.

Let's chat search syntax!

I searched for each banned term individually so we could see how many hits for each.

• gender[Text Word]: 477,056
• transgender[Text Word]: 16,278
• "pregnant person": 176
• "pregnant people": 17,567
• LGBT[Text word]: 2,584
• transsexual[Text word]: 1,373
• non-binary[Text word]: 1,416
• nonbinary[Text word]: 2,694
• "assigned male at birth": 376
• "assigned female at birth": 655
• "biologically male": 49
• "biologically female": 29

Note, the individual words have got [Text word] next to them. This makes the search engine limit results to that exact word, otherwise it automatically adds synonyms. This is particularly relevant for "gender," as PubMed includes "sex" as a synonym, which tripled the result count. Given the directive's apparent aim to eliminate gender terminology in favor of sex-based language, I maintained this distinction.

If you want to view my results, paste the following into PubMed search:

(((((((((((gender[Text Word]) OR (transgender[Text Word])) OR ("pregnant person")) OR ("pregnant people")) OR (LGBT[Text word])) OR (transsexual[Text word])) OR (non-binary[Text word])) OR (nonbinary[Text word])) OR ("assigned male at birth"[tiab:~0])) OR ("assigned female at birth"[tiab:~0])) OR ("biologically male"[tiab:~0])) OR ("biologically female"[tiab:~0])

I recommend you sort by: Most Recent for a more shuffled/representative sample of different types of articles in the first pages.

Among the half a million articles, it appears most aren't about gender at all but make references to it when discussing demographics of research participants. It is also worth considering that these results only reflect mentions in abstracts and titles, not full-text occurrences. The actual frequency in complete articles, particularly in methodology sections describing participant demographics, would likely be much higher.

It may seem that this exercise misses the forest for the trees. But my point isn't to critique the choice of banned keywords.

I believe the goal is not just to suppress LGBT research. It doesn't matter that most articles have no direct relevance to that field. This is about mandating linguistic compliance across all scientific communications. Language shapes reality. By eliminating specific terminology, the ability to precisely communicate about certain concepts becomes limited. This extends far beyond any single field of study, as demonstrated by the prevalence of these terms in general medical literature. This linguistic erasure affects everything from cancer studies to cardiovascular research, from pediatrics to geriatrics. Every scientist affected by this ban now faces a choice: comply with ideologically-driven language restrictions, or have their work suppressed.

Comment