Decentralise Medicare: A Brighter Health Future

First published 24 April 2025. Revised 18 September 2026: factual claims checked against the sources listed at the end, unsupported claims removed, corrections listed. Part 4 of 4 in the Culture of One Medicare reform series.

The argument in one paragraph. The Digital Medicare Wallet is a proposal, not a functioning system, and the first version of this article was wrong to call it “immediately viable”. The right next step is not a national rollout but a bounded pilot that separates the information mechanism from the funding mechanism, measures outcomes across the eligible population and not only among users, and pre-registers the results that would expand it or stop it. This article sets that out and corrects the first version’s list of “outcomes we can expect”, which promised savings nobody had measured.

It still signals a shift I believe in: away from credentials as the test of who may be seen first, toward capability, collaboration and systems that serve people rather than professions. Belief is not evidence. So here is the test.

What should be piloted first?

Proposal. Inside the first stage below, three arms where feasible: usual access; visibility and navigation alone; navigation plus defined direct-access funding for a specified set of musculoskeletal presentations. That separation is the whole point. It is designed to show whether any gain comes from people knowing the door exists, from removing the referral, or from both, provided the arms are randomised, adequately powered and protected from contamination. A concept sketch is not a protocol. Start where allied health supply and medical escalation capacity already exist. Test rural delivery separately rather than assuming it transfers. Include patient representatives and First Nations governance in the relevant pathways. Use outcomes for the whole eligible population, not just users: people who self-select into a new pathway are not comparable to people who do not, without adjustment. Rare harms need surveillance beyond a small pilot.

The staged sequence from the first version stands, with the promises taken out:

1. Pilot clinics with integrated allied health teams, digital infrastructure and outcome tracking.
2. Community-led trials in underserved regions, evaluated on their own terms.
3. Phased expansion only where stage one and two results meet the pre-registered thresholds.
4. National deployment tied to MBS reform and workforce strategy, if the earlier stages earn it.

Whole-system value, not suppressed use

Proposal. Judge the model on health, function, access, financial protection and equity, alongside cost. Inference. Meeting unmet need can increase expenditure. That is not failure; it is what meeting unmet need looks like. A bigger bill still needs an explanation, which is why the cost is reported gross before any offset is claimed. Hypothesis. Some targeted pathways will also reduce avoidable acute-care demand. Evidence. Pathway-level resource evidence is supportive in places: the 2026 systematic review of direct-access physiotherapy found neutral to reduced per-episode costs, at low certainty (Fischer and colleagues). Universal downstream savings are not established, and the Oregon Medicaid experiment (Taubman and colleagues, 2014) is a randomised counterexample: coverage expansion increased emergency department use by about 40%. Boundary. “More healthcare is always better” is not defensible. Necessary treatment, low-value treatment and supplier-induced demand have to be told apart, and a rise in utilisation on its own does not tell you which one you are looking at.

What could go wrong?

Each change in the model has a mechanism it relies on, a population it can help, and a way it can reverse. Writing them side by side is the honest version of the first draft’s list of benefits.

The last row does not follow from a wallet on its own. Better-linked community and hospital care needs separate, hospital-facing service design. The first version listed reduced ED and ambulance use, faster recovery and “massive administrative savings” as outcomes we could expect. They are hypotheses, and some of them may be false.

What findings would justify expansion, or stopping?

Pre-register them, with independent investigators holding the stopping rules rather than the people running the pilot.

Expand if: time to appropriate assessment falls; function outcomes are non-inferior to usual care; escalation completion is high and delayed-diagnosis events are no worse than usual care; uptake improves for lower-income, regional and First Nations groups relative to need; and per-episode public cost stays inside a pre-set band after coordination and administration are counted.

Stop or redesign if: serious harm or delayed-diagnosis events exceed a pre-set margin; uptake concentrates in groups already well served; extensions cluster with particular providers without matching outcome gains; or total public cost exceeds the band with no measurable health or equity gain to show for it.

The cost question, in a form that can be argued with

Incremental public cost equals new benefit spending, plus administration, plus coordination, plus induced downstream spending, minus displaced existing public services, minus demonstrable cash-releasing offsets. That is an analytical structure, not a cost estimate; this model has not been costed. Three distinctions keep it honest. Cost-effectiveness and affordability are different questions: a pathway can be worth its cost and still be unaffordable in a given budget year, and the reverse. Hospital capacity released is not cash saved. And new care that meets unmet need is not the same thing as private spending moved onto the public purse, or extra low-value activity; the evaluation has to estimate each separately. Do not blend Medicare benefits with all-funder primary care expenditure as if they were one available budget.

Agency with accountability

The future of care is not more bureaucracy. It is better trust, with the accounting to back it. Trusting patients to know what they need first. Trusting clinicians to deliver care at the top of their scope. Trusting teams to collaborate without coercion. Trust here is not deregulation: access based on clinical criteria rather than legacy power, oversight that is data-informed and proportionate rather than admin-choked, safety that comes from structure and accountability, and funding that follows value delivered. Every one of those is a design intent to be tested, not a property the model already has.

The Digital Medicare Wallet asks us to stop treating healthcare as a sequence of permissions. It does not abandon GPs; it aims to free them for the work that needs a doctor. It does not elevate allied health as rebellion; it recognises work that is already happening privately. It does not give patients unchecked power; it gives them structured agency. Whether it delivers any of that is the pilot’s question, and the most important missing piece of this whole series is a costed, bounded pilot specification. Not another article.

PS. This ABC 7.30 report by Adele Ferguson from 2022 on Medicare billing is a worthwhile reminder that waste and misuse are a documented risk in Medicare billing. Any new pathway needs its audit designed in from day one.

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
• 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 (this article)

What changed in this revision

• Withdrew “functional, immediately viable” and “just a reallocation”; the model is a proposal requiring a costed pilot.
• Replaced the “outcomes we can expect” list (reduced ED and ambulance use, less burnout, massive administrative savings) with labelled hypotheses, a causal model with foreseeable reversals, and pre-registered expand-or-stop criteria.
• Added a three-arm pilot design that separates the information mechanism from the funding mechanism.
• Added the evaluation table separating health benefit, equity, capacity released, cash savings and cost, and the cost identity.
• Removed “post-dualist, post-credentialist” framing.

References

1. 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.
2. Taubman SL, Allen HL, Wright BJ, Baicker K, Finkelstein AN. Medicaid increases emergency-department use: evidence from Oregon’s Health Insurance Experiment. Science 2014;343(6168):263–268. About 25,000 lottery participants over about 18 months.
3. University of Melbourne. Evaluation of the Better Access initiative – final report (executive summary). Australian Government Department of Health, 12 December 2022.
4. 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.
5. Services Australia. Services available under a GP chronic condition management plan. Page updated 30 June 2026.
6. 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.