Pressure to prioritise gender identity over sex in research and publications

[RethinkingLife]

Ciara Curran gave some very moving testimony to a House of Lords Preterm Birth Committee hearing back in 2023. She discussed the impact of desexing language and the lack of research into how this affected women.

https://www.dailymail.co.uk/health/article-13108207/Grieving-mother-gripping-teddy-memory-late-daughter-slams-dehumanising-NHS-erasure-terms-like-mother.html

https://committees.parliament.uk/oralevidence/14291/html/

Curran is one of the authors of this recent paper (open access and free to read):

Including women in research and collecting and disaggregating data on sex is an ethical imperative. However, increasingly gender identity is being prioritised over sex in data collection and language which has ethical implications. In this paper, the authors share their experiences as study participants; a health consumer advocate, patient research advisor, and lay researcher; and academic researchers of engaging with researchers, Human Research Ethics Committees (HRECs), university ethics offices, and editors and reviewers of journals regarding data collection and communication on sex and gender identity. We argue that HRECs, researchers, and publishers must carefully consider the implications of omitting data collection on sex, mandatory and universalising gender identity questions and use of desexed language. We also propose that reduced data collection and disaggregation by sex, universal imposition of gender identity, and use of desexed language in research is decreasing data quality, reducing the willingness of some to participate in research and is culturally imperialistic. Recommendations for HRECs are made and research needs in relation to sex and gender identity are outlined. Respect for women in the conduct of research requires their sex-related experiences and needs are considered and therefore that data on sex is appropriately collected and reported upon.

Munzer, M., Jameson, N., Harris, A. et al. Sex and Gender Identity: Data Collection and Language Considerations for Human Research Ethics Committees and Researchers. J Acad Ethics 23, 1399–1414 (2025). https://doi.org/10.1007/s10805-025-09605-3

https://link.springer.com/article/10.1007/s10805-025-09605-3

Good, thoughtful section on Experiences as Researchers with Pressure to Desex Language

The HoL’s hearing was Feb 2024. Apologies for misreading that.

That's OK, I hadn't seen it first time round so the link is still appreciated

You can't provide accurate data, or make the right medical decisions either as a healthy care professional or a patient if gender is the marker and not sex.
I never cease to be amazed this happens in a medical setting.

It makes all medical research completely pointless.

It also makes social science research pretty pointless. And, as with vulnerable children and medical interventions, the issue is not just that there’s some daft soft idiots out there ticking the wrong boxes, the issue is that the survey designers themselves rewrite their surveys so that the data is useless. Thank heavens for the Sullivan report but I’m not sure how much impact that had.

As a reviewer, I repeatedly ask for a disaggregation of data by sex. Or I comment on the confusion caused by the language of gender rather than sex.

But if the journal editor chooses to leave that out of my comments to the authors, there’s nothing I can do about that. I have no control over the final published article.

That review will have negligible impact because it's almost completely divorced from the reality of what the data collected is and how it is collected. As the lead author appears never to have used raw health data in any of her publications, this may not be a complete surprise.

The first big mistake is to think that it is clinical data. Outside of trials, the data used and analysed in this sort of research/statistical publications is administrative. It's derived from clinical data but it's never complete or detailed enough to base any treatment decisions upon. They should tell the analyst what happened and when but will never have enough detail to say why A was chosen instead of B. The best source of surgical data in England comes from a financial billing system. Nothing else is close but it's not remotely clinical data.

The second, related to the first, is to assume that all the data seen is objective. Some is self reported by patients each time they start a new set of treatment. And, yes, sex and gender are two of those things. Some may choose not to complete both, some will in a way that may not be consistent with the anatomy they were born with, which is why you can find a steady trickle of male patients being treated for gynaecological diseases as well as female prostate patients. Because patients have always been able to choose to report what they want to say for this (and ethnicity fwiw), to the best of my knowledge there are no (zero) adminstrative health data sets with sex and there never have been.

This hasn't stopped good research from happening before and it's continuing now.

Actually I wasn’t thinking of medical data (though obviously sex is pretty important there), but other social science surveys. We only just stopped the census from asking about gender instead of sex - and look at the mess of the explicitly gender questions. Many workplaces, including my own, don’t know how many women they employ because they haven’t asked for years. How can we measure discrimination or pay gaps? One Mumsnetter was on here recently crafting a letter to Wolverhampton council (public sector equality duty anyone?). Women are a protected characteristic, we need to measure accurately to ensure protections are effective.

Without testing the whole workforce, all HR departments have to take on trust the information their staff provides (there are cost and privacy reasons why this idea is unlikely to be justifiable/appealing to many organisations).

As they always have had to (so really rather similar to administrative health data).

Unless an organisation is extraordinarily attractive to trans men/women and/or the sex distributions are very skewed, the proportion of trans people will be very unlikely to significantly effect any estimates. If it would, I would suggest the sample size of (self declaring) women is too small to produce reliable estimates.

Numbers matter and small numbers really can skew a whole population. I am very senior and have been for well over twenty years. When I was first promoted to my current rank my employer was lauded for paying its senior women more than its senior men. There were four of us women and one was very senior indeed so averaged out our four salaries were pretty impressive.

One senior TW in a company- like Pip Bunce - means gender pay gap reporting is meaningless.

The way that women, and other protected groups, get to address our disadvantage and challenge it is through accurate data. Women count.

Now that the number of ‘women’ charged with violent offences is rocketing (because men) and suddenly we have an exponential jump in ‘women’ rapists (because men) we need to say no. These numbers are important. These crimes are not ours. These statistics do not show our wrongs.

Thanks @RethinkingLife for sharing our paper. It is nice to know that people are reading and using it!

I’m grateful to have it to cite.

Congratulations on the publication.

Cite away!

Cite away!