General Practitioner: The Protected Class

First published 22 April 2025. Revised 18 September 2026: factual claims checked against the sources listed at the end, unsupported claims removed, corrections listed. Part 2 of 4 in the Culture of One Medicare reform series.

The argument in one paragraph. General practice is a workforce under documented access pressure, and it carries two jobs: clinical judgement, and an authorisation function without which there is no Medicare rebate for the chronic-condition allied health services this series is about. The first version of this article attacked the people. This version attacks the function. What is protected in Australia is not a class of doctors but a compulsory transaction, and the right test for any compulsory transaction is whether it adds clinical value in the pathway where it is required.

What is actually protected?

A definition first, because the word gets used loosely. In this series, Medicare gatekeeping means a requirement for prior medical assessment or referral to obtain payment for a specified service. It does not mean that all Australian healthcare needs a GP’s permission. Anyone can walk into a physiotherapy clinic tomorrow and pay privately. The gate sits on the public money.

Fact. For chronic condition allied health items, the referring GP or prescribed medical practitioner decides whether the patient’s condition would benefit, and Services Australia is explicit that it is not appropriate for the allied health professional to pre-empt that decision or hand the GP a partly completed referral. The authorisation function is designed in. It is not incidental.

I have removed what the first version said about where doctors went to school, what they earn and how many days they work. It was not evidence and it was not the point. “This rule protects a function” is a claim I can defend. “Every doctor defending it acts from self-interest” is not, and I do not make it.

Is GP capacity actually scarce?

Fact. In 2024–25, 27% of Australians aged 15 and over who needed a GP delayed or did not go, and 26% of those who went judged the wait longer than acceptable (ABS Patient Experiences Survey, reported by the AIHW). More than eight in ten Australians had at least one Medicare-subsidised GP attendance in the year. Medicare paid $9.7 billion in benefits for GP attendances and $2.1 billion for allied health in 2024–25.

Inference. A workforce that a quarter of the people who needed it in 2024–25 reported delaying or not seeing is, in my view, a poor place to route a transaction that, for the low-risk presentations this series is about, exists mainly to authorise someone else’s work. The lower count of funded allied health services does not by itself prove under-use; need, case mix, eligibility and private provision all differ. It does show where the public money currently sits.

Fact, with limits. Musculoskeletal presentations are a large share of general practice work. One study of a single urban UK practice (Keavy and colleagues, Family Practice, 2023) found 21.1% of 545 reviewed consultations were musculoskeletal, three-quarters of them re-presentations of an existing problem. It is one practice in one country; I use it as an illustration of scale, not an Australian estimate. An RACGP opinion piece puts a mental health component in almost 40% of consultations in a typical week; that is the college’s own framing and I cite it as such.

When does a GP-first pathway add value?

This is the honest version of the question, and it has honest answers on both sides.

It adds value when the presentation is diagnostically uncertain, when there are red flags, when medications and comorbidities interact, or when a longitudinal relationship is doing work the patient cannot see. The continuity evidence is consistent: 18 of 22 observational studies in a 2018 systematic review associated greater continuity of doctor care with lower mortality (Pereira Gray and colleagues, BMJ Open). Gatekeeping as a system has evidence too: a 2019 systematic review of 25 studies (Sripa and colleagues, British Journal of General Practice) associated it with lower use and expenditure and better quality of care, with lower patient satisfaction and a worse cancer-survival finding as the costs. And Australian chronic disease planning can carry real clinical content: a 2024 scoping review of 17 studies in diabetes care found plans improved self-management, while also finding limited patient participation and implementation gaps.

My claim, to be tested, is that it adds little when the encounter exists to sign a form for a presentation the receiving clinician is trained and regulated to assess. That is the pathway I am arguing about. Not general practice: the compulsory preliminary consult for specified low-risk presentations.

I have also removed three sentences from the first version: that GPs lack musculoskeletal training, that they do not prescribe exercise, and that medication is their only lever. They were categorical, unmeasured and unfair. GP musculoskeletal competence varies, as every profession’s does. The argument does not need a global ranking of clinicians. It needs a test of the marginal value of a compulsory encounter, pathway by pathway.

Entry and coordination are different jobs

Inference and proposal. Coordinating a person’s care over years and authorising their first physiotherapy visit are distinguishable functions. For the Medicare-funded pathways this series is about, the referral rule ties them together. The proposal is to allow specified first-contact pathways while preserving a usual clinician or team and escalating complex care to them.

Hypothesis. Direct entry can preserve or even improve continuity, but only if reporting, responsibility and coordination are actively funded rather than assumed. Boundary. A shared record is informational continuity. It is not the same thing as a trusted longitudinal relationship or coordinated management. An app is not a care team.

Evidence. Indirect. The direct-access literature and the continuity literature exist separately. I have found no trial that has tested this combined design inside Medicare, which is why Part 4 proposes a pilot rather than a rollout.

What the direct-access evidence does and does not say

Fact, with limits. Fischer and colleagues’ 2026 systematic review (21 studies, 90,401 patients; 7 randomised trials and 14 observational studies) found direct access to physiotherapy for musculoskeletal conditions was associated with 10 to 30% fewer GP consultations across 15 studies, reduced imaging and medication use in 14, non-inferior clinical outcomes in 14, and neutral to reduced per-episode costs. Study-level certainty was moderate in two, low in twelve and very low in five. Five studies assessed safety and found no serious adverse events attributable to direct access. “No reported serious harm in five studies” is not proof of safety for rare events.

A correction on PhysioDirect. The UK PhysioDirect trial (Salisbury and colleagues, 2013, 2,249 patients) is often cited, including by me, as direct-access evidence. It was mainly a telephone-first assessment service for patients who had already been referred; only 49 of 2,256 participants self-referred. It showed equivalent physical outcomes at six months, a much shorter wait (median 7 days against 34), similar NHS costs and no adverse events detected. That supports a telephone-first reorganisation of physiotherapy access without worse outcomes. It is not a clean test of removing GP referral, and I no longer cite it as one.

None of this validates direct access for all allied health, or a national funding model. It supports specified musculoskeletal pathways, at low certainty, in other countries’ systems.

The RACGP position, stated fairly

In March 2025 the RACGP’s submission on the draft National Allied Health Workforce Strategy argued that general practice should be a main setting for allied health professionals to work in multidisciplinary teams, and that GPs should remain at the centre of care teams. The first version of this article called that a swipe at allied health. Read plainly, it is an argument for GP-centred teams. My disagreement is narrower than I first wrote it: teams are good; a compulsory entry transaction through a scarce workforce is a separate design choice, and it should be justified on its own merits. The government’s own Scope of Practice Review (November 2024) examined exactly those funding and regulatory barriers to professionals working to full scope. A recommendation is not implementation, and it is not outcome evidence, but the question is on the table without me.

Symmetric accountability

Proposal. Neither inherited status nor professional registration alone settles who should be funded to see a patient first. Require evidence proportionate to risk, not identical evidentiary thresholds for every service. Hypothesis. Transparent comparison will show which gates add value and which can be removed or redesigned. Boundary. Symmetry does not mean an untested model gets a pass because the current one is imperfect. It means the harms of delay are weighed alongside the harms of misdirection, in the same table, by the same people.

What I still think

General practice is not the problem. A compulsory transaction routed through a workforce that a quarter of the people who need it reported delaying or not seeing is, in my view, the problem. Pay GPs for judgement. Do not require their permission where it adds no clinical value. Measure both sides.

Part 3 sets out the alternative: the Digital Medicare Wallet, with the entitlement and funding claims corrected.

The series

Four articles, one argument: treat making benefits visible, deciding funding eligibility, choosing an entry clinician, coordinating care and assuring quality as separate functions. Each proposition has one home section; the others link to it.

• Part 1 · Beyond the Sick Note Economy — the access problem, and the double gate of information and authorisation
• Part 2 · General Practitioner: The Protected Class — why entry and coordination are different jobs, and when a GP-first pathway adds value (this article)
• Part 3 · Digital Medicare Wallet — the proposed model, with risk-triggered escalation and the funding claims corrected
• Part 4 · Decentralise Medicare — a bounded pilot, whole-system value, and the results that would stop it

What changed in this revision

• Removed the “class warfare” framing, the material about GPs as people rather than about the function, and a political aside. None of it was evidence for the argument.
• Removed the claims that GPs lack musculoskeletal training, do not prescribe exercise, and have medication as their only clinical lever.
• Removed the pharmaceutical-industry motive narrative. No source supported it.
• Removed “people die waiting for ambulances” and the list of downstream harms presented as consequences of the referral rule; replaced with sourced access data and labelled hypotheses.
• Restated the RACGP position from its own submission rather than characterising it.
• Qualified the musculoskeletal-prevalence figure (one UK practice) and the RACGP mental-health figure (opinion piece).
• Added the direct-access evidence with its certainty, corrected the use of PhysioDirect, and added a steelman.

References

1. Services Australia. Services available under a GP chronic condition management plan. Page updated 30 June 2026.
2. Australian Institute of Health and Welfare. General practice, allied health and other primary care services. Last updated 26 March 2026. Reports the ABS Patient Experiences Survey 2024–25 and 2024–25 MBS benefits.
3. Keavy R, Horton R, Al-Dadah O. The prevalence of musculoskeletal presentations in general practice: an epidemiological study. Family Practice 2023. One urban UK practice; 545 consultations reviewed.
4. Time to recognise GPs’ mental health contributions. newsGP (GP opinion).
5. Pereira Gray DJ, Sidaway-Lee K, White E, Thorne A, Evans PH. Continuity of care with doctors—a matter of life and death? A systematic review of continuity of care and mortality. BMJ Open 2018;8:e021161. 22 observational studies; 18 associated greater continuity with lower mortality.
6. Sripa P, Hayhoe B, Garg P, Majeed A, Greenfield G. Impact of GP gatekeeping on quality of care, and health outcomes, use, and expenditure: a systematic review. British Journal of General Practice 2019;69(682):e294–e303. 25 studies.
7. Ghasemiardekani M, Willetts G, Hood K, Cross W. The effectiveness of chronic disease management planning on self-management among patients with diabetes at general practice settings in Australia: a scoping review. BMC Primary Care 2024;25:75. 17 included articles.
8. Fischer M, Bui E, Besombes L, François M. Systematic review of direct access physiotherapy for musculoskeletal conditions in primary care: consequences for general practitioner workload, resource use, and organisation of care. BMC Primary Care 2026;27:75. doi:10.1186/s12875-026-03186-9. 21 studies, 90,401 patients; study-level GRADE certainty moderate in 2, low in 12, very low in 5.
9. Salisbury C, et al. A pragmatic randomised controlled trial of the effectiveness and cost-effectiveness of ‘PhysioDirect’ telephone assessment and advice services for physiotherapy. Health Technology Assessment 2013;17(2). 2,249 patients; 49 of 2,256 self-referred.
10. Liotta M. RACGP demands GP recognition in allied health strategy. newsGP, 5 March 2025.
11. Australian Government Department of Health and Aged Care. Unleashing the Potential of our Health Workforce – Scope of Practice Review, final report. Published 5 November 2024.