Introduction by Croakey: When considering health data and also policy responses to said data, the idiom “the devil is in the details” is useful to keep in mind.
According to Wikipedia, this expression arose from an earlier phrase “God is in the details”, “expressing the idea that whatever one does should be done thoroughly; that is, details are important”.
Apparently, the former United States House Speaker Nancy Pelosi once observed, when referring to the finer points of legislation, that “the devil and the angels are in the details”. More recently, says Wikipedia, the expressions “governing is in the details” and “the truth, if it exists, is in the details” have appeared.
All of which is to say, at Croakey we thank health policy analyst Charles Maskell-Knight for digging deep into the detail of health data in the recent Productivity Commission’s report on government services, revealing considerable variations between and within jurisdictions in service provision.
While the implications of some of those details may be debatable or unclear, at the very least they are a reminder of the peril of one-size-fits-all solutions in healthcare policy, and of the importance of understanding detail at local levels.
“Let’s hope that health departments across the country are thinking about what the results might mean for service improvement, rather than simply writing defensive briefs and hoping it all goes away,” writes Maskell-Knight.
Charles Maskell-Knight writes:
In a recent article I tried to cherry-pick the most important points from the health part of the recently released Productivity Commission 2024 Report on Government Services.
I noted in that article that many of the performance indicators reported by the Commission are also reported by SEIFA, remoteness, and First Nations status, and undertook to explore some of these results in a future article. Here it is.
Socio-economic measures
The Australian Bureau of Statistics produces four SEIFA (Socio Economic Indexes For Areas). The one used by the Productivity Commission is the IRSD (Index of Relative Socio-economic Disadvantage), which uses variables such as low income, low educational attainment, high unemployment, and jobs in relatively unskilled occupations to measure disadvantage.
At a national level, people from areas with low levels of disadvantage are marginally more likely to be seen on time in an emergency department that people from areas with high level of disadvantage.
However, looking at the data by state, in a number of jurisdictions the opposite is true. For example, for triage category 3 (urgent), Western Australia saw 36 percent of people from SEIFA 1 areas (the quintile of most disadvantaged areas) on time, compared with 32 percent for SEIFA 5 areas (the least disadvantaged quintile). For South Australia and Tasmania, the figures are respectively 45 percent compared with 31 percent, and 48 percent compared with 32 percent. A similar pattern applies across emergency department triage categories 4 (semi-urgent) and 5 (non-urgent).
At a national level, the median waiting time for elective surgery for people from SEIFA 1 is 44 days, almost half as long again as the 30 days waited by people from SEIFA 5. This disparity is particularly marked in NSW: SEIFA 1 patients wait almost twice as long (61 days) as SEIFA 5 (33 days). The pattern in other states (excluding Tasmania) is not as marked.
In my earlier article, I noted that people from SEIFA 1 were more likely to be admitted to public inpatient mental health services than people from SEIFA 5, while the opposite pattern applied to MBS funded mental health services provided in the community. As I said, either wealth is a mental health risk factor, or the distribution of private psychiatric services is highly inequitable.
Remoteness
To measure remoteness, the Productivity Commission generally uses a five-point classification: major cities, inner regional, outer regional, remote, and very remote.
The number of full-time equivalent GPs per 100,000 population in the outer regional/remote/very remote area (89) is about 75 percent of the number in both the major cities and inner regional areas.
The number of full-time equivalent public sector dentists per 100,000 population is higher in the outer regional area (5.8) and the remote/very remote area than in major cities (5.1) and the inner regional area (4.9). Given that these areas tend to have higher levels of socio-economic disadvantage, and hence more people eligible for public dental services, this is not surprising.
The data on median ambulance response times to a Code 1 incident is reported on a capital city only and state-wide basis. For all jurisdictions there was a difference of less than a minutes between the two, with the state-wide number lower than the capital city in a number of instances. In Queensland, the difference was 2.3 minutes – 14.8 minutes in Brisbane, but only 12.5 minutes across the state as a whole. This suggests that ambulance services in Queensland’s many regional centres are doing much better than in the city.
There is a consistent national pattern across emergency department triage categories 2-5: the more remote the patient’s address, the more likely they are to be seen on time. In some case the differential is quite startling – for example, in WA only 21 percent of urgent patients from major cities are seen on time, while 73 percent of very remote patients are. This may well reflect much busier emergency departments in metropolitan centres.
There was no consistent national pattern for elective surgery waiting times by remoteness. In a number of jurisdictions the median waiting time was much the same across different areas. In NSW and Tasmania waiting times increased with remoteness (NSW times increased from 44 days in major cities to 100 days for patients from very remote areas), while in SA waiting times decreased with remoteness from 65 days in major cities to 21 days in very remote).
Utilisation rates for public inpatient mental health services trended upwards with increasing remoteness, from 1.5 percent of the population in major cities to 3.4 percent in very remote areas. This presumably reflects the lack of any other treatment options such as day programs and community-based services in remote areas, meaning patients need to be provided with admitted patient care. As a corollary, utilisation rates for MBS funded mental health services decreased sharply with remoteness, from 10.9 percent of people living in major cities to 3.1 percent for people in very remote areas.
Another mental health performance indicator is the proportion of patients followed-up in the community within seven days of discharge from a psychiatric admission. Different jurisdictions have vastly different performances: only 55.2 percent of NSW patients from very remote areas receive follow-up, compared with 98.2 percent of those patients in Victoria. There are large differences within states as well: 60.0 percent of Tasmanians living in major cities receive follow-up, compared with 92.7 percent in remote areas, and 50.0 percent in very remote areas.
I have no idea what is going on here, and I suspect that mental health services may not be much better informed – but there is clearly room for some jurisdictions to learn from others.
First Nations people
At a national level, 58.4 percent of First Nations people aged 65 years and over were vaccinated against seasonal influenza, compared with 61.1 percent of non-Indigenous people. However, the rate for First Nations people was higher than for others in NSW, Tasmania, and the two territories, with the national average dragged down by Queensland, WA, and SA.
The age-standardised rate of hospital separations per 1,000 people for preventable conditions was 65.4 for First Nations people, compared with 22.0 for the rest of the population. The rate in the NT was 111.3 per 1,000.
Nationally, the proportion of First Nations people seen on time in emergency departments was marginally higher than the proportion of non-Indigenous people. This may well be because of the higher proportion of First Nations people residing in outer regional/remote/very remote areas, where emergency department performance is better than the cities and inner regional areas.
However, seven percent of First Nations people attending emergency departments left before being seen, compared with 4.3 percent for other people; and 4.1 percent left at their own risk (2.7 percent for other people).
The median waiting time for elective surgery for First Nations people is longer than for other people nationally (50 days compared with 39 days) and in every jurisdiction except SA. While some of this difference may partly reflect First Nations people in some states living in remote areas where access is worse generally, it is not improbable that some of the difference is due to racism and a lack of cultural safety.
Data on unplanned readmission to hospital with 28 days of a set of seven selected procedures shows that the readmission rate for First Nations people is substantially higher than others for hip and knee replacements, hysterectomies, prostatectomies, and appendectomies. However, the rate for tonsillectomies and adenoidectomies is lower (44.9 per 1,000 separations, compared with 48.2) and cataract surgery (1.1 compared with 2.9).
Utilisation rates for public inpatient mental health services by First Nations people are higher across all jurisdictions than the rates for non-Indigenous people. While the national utilisation rates for MBS funded mental health services are the same for First Nations people and non-Indigenous people (11.1 percent), the rates by jurisdiction differ greatly. In Victoria the rates are 16.6 percent First Nations compared with 11.9 percent non-indigenous, while in the NT the rates are 3.0 percent compared with 7.0 percent.
The proportion of patients followed-up in the community within seven days of discharge from a psychiatric admission is lower for First Nations people than non-Indigenous people nationally (71.6 percent compared with 75.8 percent) and is consistent across all jurisdictions. Again this raises issues around cultural safety and institutionalised racism.
The report includes data on the age-standardised rate of adults with high or very high levels of psychological distress.
The rate for First Nations people in 2017-19 was 31.2 percent, well over double the rate for non-Indigenous people of 13.8 percent. Given this background, it is hardly surprising that the suicide rate for First Nations people for the period 2018-22 of 27.6 per 100,000 people was also more than double the rate of 12.2 for non-Indigenous people.
What to make of all this?
The first point to bear in mind is that the three dimensions of SEIFA, remoteness, and First Nations status are related. Areas outside the metropolitan conurbations tend to have lower SEIFA, and First Nations people make up a greater proportion of the population in outer regional and remote areas. The Productivity Commission does not try to disentangle these overlapping effects, but they are clearly important.
For example, if emergency departments in outer regional areas are less busy than those in cities and are able to treat more patients in a timely way, patients from those areas – who are also more likely to be of lower SEIFA status and more likely to belong to First Nations – will have better access. Similarly, the trend for use of in-patient psychiatric services to increase with remoteness and with socioeconomic disadvantage and with First Nations status are also linked.
The second point that emerges from the data is the extent of interjurisdictional variation in some indicators.
Why do median waiting times for elective surgery in SA decrease as the remoteness of the patient’s residence increases? Why do First Nations people in Victoria have much higher rates of utilisation of MBS-funded mental health services than other people?
These outlying results may reflect random idiosyncratic features of service delivery mechanisms – or they may reflect deliberate design choices that could be adopted elsewhere. Let’s hope that health departments across the country are thinking about what the results might mean for service improvement, rather than simply writing defensive briefs and hoping it all goes away.
Finally, the data confirm the finding in the Closing the Gap Report 2023 that the gap between First Nations peoples and other Australians in “enjoying high levels of social and emotional wellbeing” is worsening.
The most recent result from 2017-19 is that 31.2 percent of First Nations people experienced high or very high levels of psychological distress, an increase from 29.4 percent in 2011-13. Following the failure of the referendum on The Voice and the toxic, racist public debate accompanying this, I am sure the rate will have increased again.
The health gaps shown in the report are only one dimension of the gap in experience between First Nations people and other Australians, and the health system alone can’t close them.
As the Productivity Commission’s other recent major review highlights, all Governments need to fundamentally change how they engage with First Nations people in Closing the Gap if real progress is to be made.
• Charles Maskell-Knight PSM was a senior public servant in the Commonwealth Department of Health for over 25 years before retiring in 2021. He worked as a senior adviser to the Aged Care Royal Commission in 2019-20. He is a member of Croakey Health Media. Follow on X/Twitter at @CharlesAndrewMK.
Further reading
• The health system is under stress, and here are some of the ways that is affecting patient care
A note from Croakey Health Media
We are seeking sponsorship for the #Medicare40Years series of webinars, and welcome approaches from organisations interested in supporting informed public debate about health reform. Contact us.
Bookmark this link for more stories on #Medicare40Years
