When a Map Moves an Area from MM2 to MM3, Who Receives More Healthcare Support?

When Numbers Start Making Decisions · Season Three, “The People the Average Does Not See” · Article 2

1. A boundary on a map can become a boundary in healthcare

Consider two general practices separated by a short drive. Both struggle to recruit clinicians, serve older patients and manage long travel times to hospitals. One is classified MM2 and the other MM3 under Australia’s Modified Monash Model. That single step can affect eligibility for workforce incentives and the size of payments intended to attract and retain health professionals.

The model classifies locations from MM1, metropolitan, to MM7, very remote. It uses population and remoteness information to distinguish large regional centres, medium and small rural towns, remote communities and very remote areas. Government programs then attach different rules to those categories. Under the Workforce Incentive Program, for example, location can affect both whether a service qualifies and the loading or payment available. Department of Health: Modified Monash Model

The map performs an essential task. Australia cannot negotiate a separate rurality definition for every clinic and every payment. Yet a classification designed to represent geographic difficulty also creates an administrative threshold. If a revised map moves a place from MM2 to MM3, the roads and patients do not change that morning, but its institutional opportunities may.

The question is therefore who the map describes, which disadvantages it captures, and how much authority a geographic category should have over a particular service.

2. Modified Monash is a proxy for the difficulty of serving a place

Remoteness is not one thing. Distance to a capital city matters, but so do town size, labour markets, nearby hospitals, transport and the capacity to support a clinician’s family. The Modified Monash Model combines geography and population size so that a large regional city is not treated as equivalent to a small town merely because both are outside a metropolitan boundary.

That is an improvement over a blunt urban-rural split. A community of 30,000 usually supports different professional networks and services from a community of 3,000. A remote town connected by a reliable regional route differs from an isolated settlement with seasonal access. Seven categories give programs a common language for graduating support.

But the model still describes a location, not an individual doctor, practice or patient. A practice on the edge of an MM2 centre may face greater recruitment difficulty than an MM3 practice near a strong regional hub. A clinician may commute across the boundary. Patients may cross it for care. A town’s population can include communities with very different access.

The map is thus a proxy for a bundle of likely disadvantages. It is useful precisely because government cannot observe the full recruitment and service conditions of every practice in real time. Its efficiency comes from replacing detailed cases with a stable geographic signal. The signal should guide attention without being mistaken for the underlying reality.

3. The MM2–MM3 line is made by rules and data versions

No physical feature marks the point at which MM2 ends and MM3 begins. The category results from definitions applied to population and remoteness data. The current 2023 Modified Monash Model replaced the 2019 version, and the government’s Health Workforce Locator allows users to check classifications by address. Health Workforce Locator

Version changes are not mere technical maintenance. Population growth can move a town into a different size category. Revised boundaries or source data can alter the result. A program that refers to one version can produce a different eligibility answer from a program using another. When the map changes, affected services need to know which date controls, whether existing participants are protected and how long transition lasts.

The closer a service lies to a boundary, the less plausible it is to treat the category as a sharp change in actual need. MM2 and MM3 are administratively distinct but geographically continuous. A transparent program should publish the model version, effective date, address used and treatment of edge cases.

Without versioning, a provider can appear to become more or less rural for no intelligible reason. With versioning, the change can be explained as a revised measurement of place rather than a sudden change in the people who live there.

4. A national map cannot see local recruitment conditions

Population and remoteness capture important pressures, but not all of them. Housing shortages, school availability, professional supervision, cultural safety, hospital vacancies, spouse employment and the sustainability of on-call rosters can decide whether a clinician stays. A fire, flood or mine closure can change conditions faster than the classification data.

The same MM category can also contain communities with very different health burdens. First Nations communities may face barriers not represented by town size. A practice serving a dispersed population may carry travel and outreach demands unlike another at the same classification. Seasonal tourism can multiply patient numbers without appearing in usual resident population.

This does not make the model useless. It identifies systematic patterns and provides a defensible national baseline. The problem begins when eligibility generated from the baseline is presented as a complete assessment of local workforce need.

Programs can reduce that error by combining the map with vacancy duration, turnover, service volume, travel time, population health and evidence from local organisations. A standard map is strongest as the first layer of allocation. The more severe the consequence of exclusion, the stronger the case for a second layer that can hear local evidence.

5. Who benefits when the classification rises?

An MM3 classification does not itself deliver a clinician. It can make a practice eligible for incentives or increase a payment, but the path from eligibility to patient care has several steps. The practice must participate, recruit or retain eligible professionals, meet program requirements and translate the workforce benefit into available services.

The Workforce Incentive Program has different streams and detailed rules. In the Practice Stream, rural loadings vary by Modified Monash category. In the Doctor Stream, eligible locations, activity periods and years of service affect payments. Workforce Incentive Program: Practice Stream incentive payments Doctor Stream payment amounts

The category therefore interacts with professional status, billing, service history and program design. Two clinics in the same MM area may receive different benefits. Patients within the area may still experience different access according to transport, cost, disability, language and appointment availability.

Evaluation should follow the chain beyond payments. Did vacancies shorten? Did turnover fall? Were appointments added? Which groups used them? A map can allocate a workforce incentive efficiently while the intended patients remain unseen. Classification accuracy is not outcome evidence.

6. Transition rules are part of fairness

When a new model moves an area to a less favourable category, immediate withdrawal can destabilise a service that made staffing commitments under the old rule. When it moves an area to a more favourable category, long delays can perpetuate an acknowledged mismatch. Fairness therefore requires explicit transition.

Good transition rules state which version applies, preserve commitments for a defined period, prevent retrospective repayment caused only by reclassification, and communicate the change early. They also avoid making temporary protection permanent when current data show that resources should move elsewhere.

Appeal should be equally clear. A provider must be able to correct a wrongly geocoded address or misapplied version. That is different from asking officials to redraw the national model because a practice has exceptional difficulty. The first is an error correction; the second is a request for a program-level exception or a separate needs mechanism. Keeping them distinct makes reasons and responsibilities visible.

The more a program depends on a fixed boundary, the more it should publish anomalies and monitor bunching near that boundary. Otherwise, lobbying over classification may replace discussion of the underlying workforce problem.

7. A practical framework for map-based support

For any program using the Modified Monash Model, ask:

1. Which model version and exact address determine the result?
2. Is the program trying to support a location, a provider, a clinician or a patient?
3. Which aspect of rural difficulty does the category represent?
4. What relevant conditions does it omit?
5. How different are places immediately across the boundary?
6. Can data and geocoding errors be corrected promptly?
7. Is there a separate route for exceptional local need?
8. Do transition rules protect reasonable reliance without freezing old allocations?
9. Are workforce and patient outcomes audited after payments are made?

This framework does not demand individual negotiation for every location. It demands that a standard classification remain connected to the problem it was designed to approximate.

Conclusion: the map should organise support, not settle the whole case

The Modified Monash Model is valuable because it makes geographic workforce policy possible at national scale. It recognises degrees of rurality that a metropolitan/non-metropolitan split would erase and creates consistent rules across programs.

Its categories are nevertheless representations. An MM2–MM3 boundary can change institutional treatment more sharply than local conditions change. My judgement is that the map should establish a transparent baseline, while versioning, transition, error correction and evidence of exceptional need limit its authority. Programs must then measure whether more professionals and accessible care actually reach the community.

A map can reveal spatial disadvantage. It cannot see every clinic, clinician or patient contained inside its colours.