Trial access equity by state

Recruiting trials per million residents

Recruiting trials with at least one site in each state or territory, divided by that state's population. A larger state having more trials is unremarkable; this asks whether it has more trials for its size.

These rates compare states with each other, not with a national figure. Across Australia there are 910 distinct recruiting trials with an Australian site, or 32.7 per million people — a lower rate than any single state above. That is not a contradiction and not an error. A trial recruiting in both Melbourne and Sydney is genuinely available in both states, so it counts once for Victoria and once for New South Wales, but only once for Australia. Each state's rate is a fair answer to “how much trial access does a resident here have”, and the state rates are built identically so they can be read against each other. None of them can be read against the national rate, added together, or averaged.

The smallest jurisdictions rest on very few trials. Northern Territory's rate comes from 6 recruiting trials, so a single trial opening or closing moves it by around 17%. Treat the bottom of this chart as a rough signal rather than a precise measurement, and expect it to move between updates in a way the larger states will not.

62 recruiting trials list an Australian location with no state or postcode we can resolve, so they are not counted in any state's numerator above. Every state's rate is therefore slightly understated. We would rather leave those trials out than guess a state you might travel interstate on.

These counts come from ClinicalTrials.gov only. Trials registered solely with the Australian New Zealand Clinical Trials Registry (ANZCTR) are not included, so every state's rate understates actual local trial access.

Population figures are the ABS Estimated Resident Population for 2025-Q4, the most recent quarter published. Trial counts refresh nightly, so the two sides of this calculation are not from the same day: the population denominator lags by a few months, which nudges every rate slightly high as the population grows.

Population: Australian Bureau of Statistics, Estimated Resident Population (2025-Q4), used under CC BY 4.0. Trial counts: ClinicalTrials.gov registry records, refreshed nightly. The ABS national total of 27,801,023 includes residents of the Other Territories, who are counted nationally but in no state, which is why the eight states above do not add to it. Methodology: /insights/methodology/. This page does not constitute medical advice.

Where trial sites sit on the socioeconomic scale

Every Australian trial site has a postcode, and the ABS scores each postcode for socioeconomic advantage. This asks where the sites are — not who takes part in the trials run there.

Trial sites are concentrated in the most advantaged postcodes, and it is not simply because more people live there. The most advantaged tenth of postcodes holds 15.5% of Australians but 34.1% of trial sites — 2.20 times its share. The three most disadvantaged deciles together hold 21.3% of Australians and 4.5% of trial sites. Comparing each decile against its own population is the whole point of the second bar: the advantaged deciles genuinely do contain more people, and the gap remains after allowing for that.

This describes areas, not people, and it describes hospitals, not patients. Two things follow. First, a SEIFA score is a smoothed statistic about a postcode as a whole: plenty of people on low incomes live in advantaged postcodes and plenty of well-off people live in disadvantaged ones, so nothing here tells you about any individual. Second, and more important: a hospital's catchment is far wider than its own postcode. Australia's big teaching hospitals sit in inner-city postcodes that are now among the most advantaged in the country, and they treat patients from across their whole city and often from interstate. So the honest reading of this chart is where the hospitals that run trials happen to be, which is largely a matter of where they were built. It is not evidence that disadvantaged Australians are excluded from trials, and we would need data on actual participants — which no public registry provides — to say anything about that.

The largest research precincts are missing from this chart, which probably understates the concentration. ABS postal areas approximate residential postcodes, so they exclude the codes assigned to institutions rather than neighbourhoods. Those are exactly the big academic centres: Parkville, Herston, Concord and Macquarie Park between them host hundreds of trial site listings and have no SEIFA score at all. 229 of 3,765 Australian site listings drop out this way, and a further 437 list no usable postcode. Those precincts are inner-city and would almost certainly fall in the higher deciles, so including them would most likely make the pattern above stronger, not weaker.

791 of the 910 recruiting trials with an Australian site (87%) have at least one site we can place in a decile; 119 have none. The chart counts site listings rather than trials, because each listing has exactly one postcode and therefore one decile, which makes the shares add to 100%. Trials cannot be counted that way: a trial running in a decile 3 postcode and a decile 9 postcode belongs to both.

Socioeconomic scores: Australian Bureau of Statistics, Socio-Economic Indexes for Areas (SEIFA) 2021, Index of Relative Socio-economic Advantage and Disadvantage by Postal Area, used under CC BY 4.0. Population shares here are 2021 Census counts, which is what SEIFA is built from — they are a different measure from the Estimated Resident Population used earlier on this page and the two are not mixed in any calculation. Trial sites: ClinicalTrials.gov registry records, refreshed nightly. Methodology: /insights/methodology/. This page does not constitute medical advice.