Unit cost is the number a program of this kind stands or falls on, and the number a department has the strongest incentive to leave out of an evaluation.

It is in this one, and in more detail than a summary conveys. The second interim report estimates Commonwealth funding at $236 per presentation across the 53 Tranche 1 clinics where a patient would otherwise have attended an emergency department, against $617 for the equivalent emergency department presentation, leaving a saving of $381. It then publishes two different estimates of how many presentations were actually avoided -- about 260,000 on the program's own module data, about 95,000 on a difference-in-differences analysis of hospital data -- and carries both through to annual savings of $99 million and $36.2 million respectively, the latter with a confidence interval of $25.4 to $48.1 million.

Publishing a second method that produces roughly a third of the headline saving, and an appendix setting out how the unit cost was calculated, is the behaviour this register is trying to identify.

A reader can therefore form their own view of whether the program is worth its cost, which is the whole purpose of publishing an evaluation.

Corrected from an earlier draft that asserted cost per presentation was the one number not published. It is the headline finding of the second interim report, and the assertion was made without opening it.

The register keeps one real limitation against this entry rather than pretending the reporting is complete. Data is at program level, so per-clinic utilisation and cost are not available. At small clinics that would raise a genuine identifiable concern about patients, which is a constraint rather than an excuse -- and the answerable version of the finer question is a suppressed-cell release, not raw per-clinic data.

The ANAO has separately examined whether the program's performance measures were established well enough to tell if clinics achieve their intended outcomes, which is a different question from whether the cost is published.

Opened and read on 2026-09-03, and the figures in this entry were corrected as a result. It previously stated $246.50 per presentation, taken from reporting about the evaluation rather than from it. That figure does not appear in the second interim report. The report gives $236 for ED-substituting presentations, notes a decrease against a $216 figure in the first report on a different basis, and attributes $134.10 of the per-presentation cost to grants.

The $617 comparator was correct, but was described here as the saving. It is the cost of the emergency department presentation being avoided; the saving is $381.

A small entry, three wrong numbers, all from summaries. The correction is the argument for the verified standard in one example.