We have a name for prescribing by nurses, pharmacists, podiatrists and optometrists: non-medical prescribing. We do not have a corresponding term for prescribing by doctors. That is simply called prescribing.
The language tells us something before the evidence does. Medicine is the default. Everyone else is the exception.
When a nurse practitioner, pharmacist or optometrist prescribes, the burden of proof falls on them. They must demonstrate safety, competence and equivalence. When a doctor prescribes, the authority is assumed. The question is not asked.
That asymmetry is not primarily the product of clinical evidence. It is the product of institutional inheritance — and of a classification system that reproduces that inheritance every time it is used.
How medicine became the default
Medical ownership of prescribing was not established by a single legislative act. It accumulated across more than a century of institutional decisions, each of which reinforced the others.
The early systems of medical registration — the 1858 Medical Act in Britain, analogous laws in Australian colonies — gave the medical profession not a simple monopoly but something more durable: the mechanisms through which the roles of other emerging health occupations could be shaped and restricted. Medical practitioners sat on the registration boards of other health professions well into the late twentieth century. Access to many allied health services was made conditional on medical referral. Funding arrangements were built around medical gatekeeping as their default architecture.
Prescribing authority sat at the centre of this structure — medicine’s most visible jurisdictional claim, reinforced simultaneously by regulation, by funding, by professional hierarchy, and by training systems that positioned the prescribing decision as the apex of clinical authority.
Medicine did not need to keep reasserting its ownership of prescribing. The institutions did it automatically. The default does not need to explain itself. The exception does.
That is what “non-medical prescribing” names. Not a category of clinical work. A category of departure from the assumed norm.
What the evidence establishes — and what it doesn’t
I use the established term non-medical prescribing because it is the language of the literature — but the term is itself the object of analysis. It defines a large and diverse group of practitioners not by what they are, but by what they are not: doctors.
The United Kingdom has had formal non-medical prescribing arrangements for decades. Australia has progressively extended prescribing authority to a range of health professions through different legislative and funding pathways. The research base across both systems is now substantial. Decades of evidence have established that appropriately trained practitioners can prescribe safely within defined scopes of practice and governance arrangements.
What that evidence does not establish — and does not need to — is that every practitioner should prescribe every medicine in every context. That would walk straight back into the “who does it better?” trap: a jurisdictional contest framed as a clinical question.
The evidence no longer supports treating prescribing by appropriately trained non-medical practitioners as inherently unsafe. What remains genuinely contested is how far prescribing authority should extend, to whom, for which medicines and conditions, and under what governance and funding arrangements.
That is a different and more tractable question.
Prescribing is not one thing
Part of what makes the debate so persistent is that “prescribing” is treated as though it were a single, indivisible capability — the kind of capability only a medical training can produce.
It is not.
Writing a repeat prescription for a stable chronic condition is not the same work as initiating a high-risk medication. Managing polypharmacy in a medically complex patient is not the same work as prescribing within a narrow, defined specialist scope. Diagnosing an undifferentiated presentation and deciding whether medication is indicated at all is not the same work as titrating a medication within an established protocol for a known condition.
Once prescribing is treated as a single capability, the professional argument almost automatically wins: who has the broadest training? But that may be the wrong unit of analysis altogether.
The better question is: what prescribing decision is being made, for which condition, at what level of clinical complexity and uncertainty, and what does that specific work require? That question can actually be answered. It is also the question that the evidence — where it has been gathered carefully — consistently addresses. Pharmacists prescribing within their specialist scope of medicines management. Nurse practitioners prescribing for chronic conditions they manage longitudinally. Optometrists prescribing for ocular conditions that fall squarely within their clinical expertise. In each case, the safety question is tractable precisely because the work has been specified rather than aggregated.
Why extending prescribing is harder than it looks
If the evidence is reasonably clear, why does every extension of prescribing rights require the same arguments to be made again from the beginning?
Because evidence addresses only one part of the problem. A regulatory change that extends prescribing authority does not, by itself, change the funding arrangements that make the role economically viable. It does not change the education and training pipelines that produce prescribers at scale. It does not change the industrial and workforce classifications that determine how employing organisations deploy and pay those practitioners. And it does not change the professional culture that has, for a century, positioned prescribing as medicine’s apex skill.
I call this professional boundary instantiation: professional authority is not located in one law or rule. It is embedded across the institutions that educate, regulate, classify and fund work. Reform one mechanism and the others can absorb the change. The reform announces itself, creates some activity at the point of change, and is quietly neutralised by the mechanisms it did not reach.
Pharmacist vaccination offers a partial counter-case: when several mechanisms began moving in the same direction — regulatory authorisation, public funding, training pathways and public health messaging — uptake expanded rapidly. The lesson is not that all four boxes must be ticked simultaneously. It is that the more mechanisms that move together, the more durable the change becomes.
That is why piecemeal reform rarely sticks, and why the debate about prescribing rights keeps returning to the same starting position. Each extension is won on safety grounds, justified through evidence, and then partially absorbed by the institutional architecture that surrounds it. The evidence accumulates. The burden of proof barely moves.
Prescribing is work, not a profession
The argument is not that professional background is irrelevant to prescribing. It is that professional title alone does not tell us what prescribing work is being performed, how complex it is, or whether the practitioner doing it has the capability the work requires.
A medical degree is not a homogeneous credential. Doctors differ enormously in their prescribing competence across different therapeutic areas. General practitioners prescribing in areas of specialist complexity cannot be assumed to be safer solely by virtue of professional title than appropriately trained prescribers working within an established specialist scope. The profession does not tell us enough about the work. The work has to be examined on its own terms.
This matters because it shifts the question. Not: is non-medical prescribing safe? But: is this practitioner, doing this prescribing work, at this level of complexity, under this governance framework, safe and effective? The first question is a jurisdictional question wearing clinical clothes. The second is the question the evidence can actually answer.
The hardest professional boundaries to see are the ones that no longer need defending. They have become the default. Doctors prescribe. Everyone else must explain why they should be allowed to.
Until we ask why the burden of proof falls so unevenly — and recognise that the asymmetry is institutional rather than clinical — we will keep mistaking an inheritance for a fact.
Susan Nancarrow is a health workforce sociologist and Principal of HealthWorX Futures. She is co-author of The Allied Health Professions: A Sociological Perspective (Policy Press, 2021).
