Health Grid Platform · Australia

Every therapeutic area. Every company. One subscription.

Health Grid is a subscription analytics platform for the Australian market: pharmaceutical claims, medical services claims, hospital activity and hospital diagnoses, in one place, refreshed every month. Track a molecule, a service, a therapy area or a company nationally or state by state, and compare any of it month on month or year on year. Hospital activity goes down to individual hospitals. No commissioned study, no analyst queue, no training course. Log in and start.

5modules, drug claims, services claims, hospital insights, diagnosis overview, explorer
Monthlyrefresh, with newly listed codes captured automatically
100%of figures traceable to the source file they came from
Why this, and not the alternatives

A commissioned study answers one question. This answers the next one too.

A commissioned studyHealth Grid Platform
ScopeOne indication, one projectEvery therapeutic area and every company, at once
GeographyUsually national onlyNational and state, on every claims view
Time to an answerWeeks to monthsMinutes, for anything the data already covers
When a new product listsYour study is now out of datePicked up in the next monthly refresh, automatically
Who can use itWhoever commissioned itAnyone on your team with a licence
Cost shapeA programme of workA subscription

Five things it does that a commissioned study cannot

01

The whole market, not a slice of it

Not one indication and one competitor. Every therapeutic area and every company on the Australian schedules, side by side, all the time. The comparison you did not think to ask for is already there.

02

It keeps up with the market on its own

New products and services added to the Australian schedules are picked up in the monthly refresh. You do not commission an update; the update happens.

03

It does not need a project to be worth having

Most questions in a commercial team are small and urgent. They do not justify a study, and they do not survive a queue. This answers them the same afternoon.

04

Nobody needs training

Pick, click, read. If a colleague can use a spreadsheet filter, they can use this. There is no onboarding programme, because there is nothing to onboard.

05

You can make it yours

Build your own market baskets from item codes or whole molecules, name them, save them. Every monthly refresh flows into them automatically. Save views, export what you need, and produce your own mini-report without asking anyone.

It is priced as a subscription, not as a programme of work. Which is the point: a standing capability your whole team can use should not cost what a single commissioned study costs. How it is licensed is deliberately flexible, by named user or across the enterprise, all modules or only the ones you want. Pricing on request

The modules

Five modules. Licence the ones you need.

Access is scoped per module and enforced server-side at request time, so a user licensed for one module cannot reach another's data.

1Drug Claims

Every subsidised script, by molecule, brand, item code and sponsor.

Market Overview · Market Dynamics · Geography · State Trend · Script type · Price and Erosion · Growth · Company · My Portfolio · Item Directory

1,307 molecules, 12,178 item codes, 3,643 brands, 452 therapeutic classes. National and state, complete from 1992.

2Services Claims

The demand signal upstream of the prescription, with age and sex.

Market Overview · Group Composition · Geography · State Trend · Demographics · Fee per Service · Growth · Item Directory

5,784 items across 8 categories, 91 groups and 323 sub-groups. National series from 1993, age by sex from 2016.

3Hospital Insights

Which named facility performs the procedure, at what volume, against which peers.

Overview · Detail · Geography · Map · Segmentation · Directory · Methodology

1,165 facilities, 929 carrying published measures, and 152 intended procedures grouped into 17 therapy areas.

4Diagnosis Overview

Thirteen years of what Australia is actually admitted for.

Overview · Detail · Demographics · Directory · Methodology

22 ICD chapters and 1,547 diagnosis codes, 2011-12 to 2023-24, with same-day and overnight splits and age by sex. National level, stated on screen.

5Explorer

Any drug codes, any service items and one hospital measure, on one screen.

Aligned state heatmaps · trend charts · per-code table

Population-level only, with no patient linkage. An overlay is not a causal claim, and the product says so on the page rather than in a footnote.

Licensed separately

Take the modules you need, add others when a question requires them.

Enforcement is server-side at request time. A client without a module cannot fetch a single file from it. Not a hidden tab, a hard block.

Per named user or across the enterprise. Annual by default, and open to discussion.

Why you can trust the numbers

Every figure traces back to the source file it came from.

  • One source of truth. Every displayed number is derived from a published source file, and nothing is modelled in without being labelled as modelled.
  • Restatements are handled properly. Publishers revise recent months. We keep every version and always use the latest, rather than adding revisions together, which is a mistake that silently inflates totals.
  • New codes are captured automatically. Each refresh compares the live schedules against our list and pulls in newly listed codes.
  • A 21-point automated check runs at the end of every refresh. If completeness or integrity fails, the refresh fails and the data does not ship.
  • Sampled audits each release. Randomly selected displayed figures are traced back to the source file and the result is logged.
  • Quarantine, never silent drops. Records that cannot be parsed are counted and reported rather than quietly discarded.

Benefit is the subsidy paid, not sales. It is never described as sales or market value, because it is not either of those things. Where a figure is an estimate, for example where an item code is shared by several sponsors, the product says so on screen rather than in a footnote nobody reads.

What Health Grid Platform is not. It is not patient-level data, and there are no patients in it. Health Grid Platform is built from published aggregates and contains no patient records at all, which is exactly why it can be a subscription rather than a governed project.

Growth · disclosure line
Growth view header showing the on-screen disclosure of filters and volume floor
Licensing and terms

How to license it.

LicensingFlexible, and agreed with you rather than imposed. Per named user or across the enterprise. All five modules, or only the ones you are interested in. Annual by default. Tell us how your team is shaped and we will price it that way.
Access controlEach user's access is scoped to the modules on their licence, enforced server-side at request time rather than hidden in the interface.
What you can exportViews, charts and tabulated exports of the views you are licensed for, with item descriptions and full context columns. Not the assembled base data, under any licence.
Personal data involvedFor your users: name and work email, for authentication and licence scoping. For patients: none, because there are no patients in the product.
Security postureWe hold no security certification and we do not claim one. In this product's case the point is close to trivial: Health Grid Platform is built from published aggregates and contains no patient records at all.
TermsTerms of service and end-user licence agreement, plus a plain-English data-use summary, published on the product and linked from its footer.
Trial or pilotWorth asking about. A short evaluation on your own therapy area tells you more than any amount of reading, and we would rather you decided that way.

See it with your own therapy area in it.

A demo built on the market you actually work in is worth more than a generic one, tell us which and we will use it.

De-identified patient-level data and publicly available data, through local alliance partners. De-identification at source. Analysis in-market, under local law. Never identifiable records.