Has an attempt to maximise inclusivity resulted in disproportionate administrative complexity, and has anybody actually demonstrated that a long predefined taxonomy produces better patient care than a small number of common choices plus a free-text option?
Modern healthcare rightly places dignity, respect and individualised care at the centre of the relationship between patient and clinician. Few would seriously argue that deliberately addressing somebody in a manner they find distressing contributes anything useful to their treatment.
There is, however, an important distinction between being inclusive and attempting to catalogue every conceivable variation of human identity within an administrative system.
That distinction raises a question that healthcare organisations should be perfectly comfortable asking:
At what point does reasonable accommodation become disproportionate administrative complexity?
A recent example can be found in an online patient-record form used by Frimley Health NHS Foundation Trust. Its pronoun selection does not simply offer he/him, she/her and they/them. The list visible in the patient interface extends through numerous alternatives including co/co/cos, e/em/eirs, en/en/ens, ey/em/eirs, fae/faer/faers, ne/nem/nirs, sie/hir/hirs, ve/vir/vis, xe/xem/xyrs, yo/yo/yos and several variants beginning with ze. It additionally provides options including any/all, unknown, choose not to disclose and name as pronoun.
The existence of such a list deserves examination—not because patients should be ridiculed for their personal choices, but because public healthcare systems should be able to justify the complexity they introduce.
The principle of inclusivity is sound
There is an entirely reasonable argument for recording how somebody wishes to be addressed.
Healthcare depends upon communication. Patients routinely disclose deeply personal information and submit themselves to examinations and procedures in circumstances where they may already feel vulnerable. Courtesy matters.
The NHS also operates within substantial equality and accessibility obligations. NHS England’s current Accessible Information Standard requires services to identify, record, flag, share, meet and review relevant communication needs. Importantly, NHS England also says that systems should make people’s needs easy to record.
That last principle deserves greater attention.
Inclusivity and simplicity are not necessarily opponents.
A system can accommodate an enormous range of people without presenting every user with an enormous range of predetermined classifications.
The problem of taxonomy
Consider what happens when an organisation attempts to solve an inclusivity problem through taxonomy.
Initially there may be:
He/him — She/her — They/them.
Someone then identifies another preference that isn’t represented. So another category is created.
Another preference subsequently emerges. Another category follows.
The process can theoretically continue indefinitely because language is not static and there is no finite authoritative list of all pronouns an individual might choose to use.
This creates an interesting administrative paradox.
The harder an organisation tries to construct an exhaustive list, the more likely that list is eventually to become incomplete.
A patient whose preferred pronoun isn’t among twenty alternatives is no more represented by the system than somebody whose preference isn’t among four.
There is a remarkably simple solution:
He/him | She/her | They/them | Other – please specify | Prefer not to say
The fourth option makes the system effectively unlimited.
Rather than the NHS attempting to predict every possible answer, the patient supplies the answer.
That is arguably more inclusive rather than less.
Where is the evidence?
This leads to the more important question.
What measurable improvement in healthcare is produced by an extensive predefined pronoun taxonomy?
That is not a rhetorical dismissal of pronouns. It is the sort of question healthcare routinely asks about almost everything else.
If we introduce a clinical intervention, we expect evidence.
If we redesign a patient pathway, we examine outcomes.
If we change documentation, we consider information governance and clinical safety.
If we introduce another mandatory field into an electronic system, we should consider its effect on workload and data quality.
NHS England’s own current guidance explicitly states that IT systems recording communication needs should take account of clinical safety risks, while its implementation guidance says the initial questions used to identify communication needs should be easy enough for all service users to understand and respond to.
Those are sensible human-factors principles.
I could find NHS guidance supporting the recording and accommodation of people’s communication needs. I could also find NHS requirements concerning accessibility and individualised communication.
What I did not find in the NHS material reviewed was evidence demonstrating that presenting patients with a long predefined catalogue of uncommon pronouns produces superior clinical outcomes, patient safety or communication compared with providing the common selections and an unrestricted “other/prefer to self-describe” option.
That distinction matters.
Evidence for treating people respectfully is not automatically evidence for a particular piece of software design.
Inclusivity should itself be subjected to proportionality
There is sometimes an understandable reluctance to question policies associated with equality or inclusion because criticism can easily be interpreted as criticism of the people those policies seek to accommodate.
The two things are not synonymous.
One can entirely support a patient’s right to say:
“Please address me in this way.”
while simultaneously questioning whether a national healthcare organisation requires twenty tick boxes to achieve it.
Indeed, proportionality is already embedded in the philosophy surrounding accessibility. The Equality Act framework concerns reasonable adjustments, while NHS England describes an anticipatory responsibility to accommodate relevant needs.
“Reasonable” is an important word.
It implies that competing considerations can legitimately be weighed against one another.
There is also a human-factors problem
Healthcare professionals already operate within extraordinarily information-dense environments.
Electronic patient records contain allergies, medications, observations, safeguarding information, diagnoses, investigation results, resuscitation decisions, communication requirements and countless administrative fields.
Every additional category has a cognitive and administrative cost, even if individually that cost appears trivial.
Good interface design therefore tries to reduce unnecessary complexity.
The objective should be to capture the minimum information necessary to achieve the intended purpose accurately.
If the purpose of the pronoun field is:
How would you like staff to refer to you?
then the database does not necessarily need to classify the patient’s answer beforehand.
It merely needs to record it.
This distinction between classification and communication is fundamental.
Sex, gender and pronouns are also different data
There is another reason why healthcare information systems require particularly careful design.
A pronoun is principally a matter of communication.
Gender identity is demographic/personal information.
Sex can, in particular circumstances, constitute clinically relevant information.
They are therefore not interchangeable concepts.
A clinician may need particular biological information when interpreting investigations, assessing disease risk, considering pregnancy, calculating medication, evaluating symptoms or planning treatment. None of that prevents the same clinician from addressing the patient using whatever reasonable form of address that patient requests.
An intelligently designed record should therefore be capable of saying, in effect:
This is the clinically relevant biological information, and this is how this person wishes to be addressed.
There need be no conflict between the two.
Problems arise when organisations attempt to make one category perform the function of another.
Inclusivity can become less inclusive
There is another unintended consequence worth considering.
Extremely elaborate inclusivity systems can actually alienate the majority of the population they are intended to serve.
A patient confronted with unfamiliar terms such as fae/faer/faers or xe/xem/xyrs may simply have no idea what they mean.
That doesn’t make the patient intolerant.
They may simply never have encountered the terminology.
NHS England itself advocates clear and accessible communication, including language that patients can readily understand.
Healthcare communication should therefore accommodate both the person requesting an uncommon form of address and the person who has never encountered it before.
Inclusivity cannot logically mean making one group comfortable by making another group afraid to admit that they don’t understand the terminology.
A simpler alternative
A proportionate system could be remarkably straightforward:
Preferred name: Richard
Pronouns:
He/him
She/her
They/them
Other — please specify
Prefer not to say
If somebody enters xe/xem, the system records xe/xem.
If somebody enters fae/faer, it records fae/faer.
If terminology changes ten years from now, the database doesn’t require redesigning.
And if 99 patients choose one of the common options while the hundredth chooses something entirely different, all 100 have still been accommodated.
That is not exclusion.
It is efficient inclusive design.
The wider question for the NHS
This example points towards something considerably larger than pronouns.
The NHS is a healthcare system operating with finite money, finite staff and finite time.
Every administrative initiative therefore carries an opportunity cost.
A minute spent completing unnecessary documentation is a minute unavailable somewhere else. A software modification has a development cost. Training has a cost. Governance has a cost. Data maintenance has a cost.
None of those arguments means that minority needs should simply be ignored because relatively few people have them. Healthcare routinely—and rightly—provides expensive accommodations and treatments for small populations.
But it does mean that proportionality remains legitimate.
The appropriate question isn’t:
“How many people want this?”
It is:
“What is the simplest effective method of meeting their legitimate need?”
That is a very different philosophy.
Common sense is not the enemy of inclusion
The debate surrounding inclusivity has become unnecessarily polarised.
One side can sometimes treat almost any accommodation as bureaucratic political correctness. The other can sometimes regard questioning the proportionality of an inclusion initiative as evidence of intolerance.
Neither position is particularly useful.
There is a substantial middle ground:
Treat everybody with dignity.
Accommodate reasonable individual preferences.
Protect clinically relevant information.
Design systems around evidence and human factors.
Keep administration as simple as the objective permits.
And, importantly:
Allow people to question whether a policy achieves its purpose efficiently without assuming that questioning the policy constitutes hostility towards the people it seeks to protect.
Conclusion
The pronoun list is ultimately a small example of a much larger question facing modern public institutions.
Inclusivity is a worthwhile objective.
But maximum complexity is not evidence of maximum inclusivity.
A healthcare system does not demonstrate compassion by accumulating categories. It demonstrates compassion by understanding what an individual patient needs and responding appropriately.
If he/him, she/her, they/them, other/self-describe, and prefer not to say can accommodate essentially the same range of patients as a continuously expanding predefined taxonomy, then the burden of proof should arguably fall upon the more complicated system.
What additional benefit does it produce?
Does it improve communication?
Does it improve patient experience?
Does it reduce discrimination?
Does it improve clinical outcomes?
And does it achieve any of those things more effectively than the simpler alternative?
If those questions cannot be answered with evidence, questioning the complexity is not an attack on inclusivity.
It is precisely what evidence-based healthcare ought to do.
The NHS should be inclusive enough to accommodate difference, confident enough to acknowledge biological and clinical reality, and pragmatic enough to recognise that good system design often means making complicated things simpler—not making simple things complicated.
