Op-ed: the Productivity Commission inquiry into mental health is a missed opportunity

By Jo-An Atkinson & Ian B. Hickie

Monday November 18, 2019

Getty Images

In the words of Benjamin Franklin; ‘Experience keeps a dear school…’ This is indeed true of mental health policy in Australia, and public policy more broadly. Despite decades of successive statutory inquiries,1-3 system reforms, significant investments by national and state governments, and substantial contributions from business, community and philanthropic organisations, rates of mental illness in Australia are not decreasing.4 There is little agreement regarding the reasons for this inertia or the appropriate strategies now required to address the complex, persistent problem of mental illness in this country.5

A range of perspectives has been offered (Box 1) many of which were reiterated in submissions to the latest Productivity Commission’s 2019 Inquiry into Mental Health. However, the draft report, released in late October, calls yet again for further ‘major system reforms,’ curiously omitting from its recommendations the proposed vital advanced decision-analytic infrastructure capable of navigating the complexity that has prevented progress thus far.

While we commend the Commission’s proposed retention of the regional focus, local engagement, and investment in areas beyond health, we are disappointed with the ongoing dependence on an antiquated decision-support framework. Without tools to make sense of complexity, decision making in mental health has often relied on an inadequate mixture of historical precedent, best guess and trial and error. While the report emphasises the importance of prioritizing interventions that have a sound evidence base, and the strengthening of data systems to support monitoring and evaluation, this doesn’t go far enough. Current tools for synthesising and operationalising research evidence are not able to answer vital questions of what are the ideal selection, design, targeting, timing, intensity, consistency, and coordination of integrated programs and services in a given context that will deliver the greatest impacts on mental health outcomes and suicide prevention.6

There are extraordinary inefficiencies in an approach that guesses at these questions, implements them, evaluates them to have been ineffective, and repeats the process iteratively; learning very little about system structure and behaviour that drive system performance, quality of care, and patient outcomes. This wasteful approach would be unfathomable in sectors such as engineering and business; yet despite lives being at stake, and despite population health and well-being being intrinsically linked to our national productivity and wealth, there remains a reluctance to evolve and invest in next-generation decision support tools that will bring a necessary discipline to new national and regional investments in transforming mental health systems to deliver better outcomes.

Systems modelling and simulation (computer simulation) provides the ability to forecast the likely impact of investments and determine the viability and comparative effectiveness of alternative strategies before implementing them in the real world. Computer simulation has long been used to solve complex strategic and operational problems, optimise system design and resource management, and improve efficiency and public safety,7 as well as having contributed to scientific and industrial advances. Computer simulation has been instrumental in achieving advanced technological exploits such as putting a human on the moon and is essential embedded infrastructure in weather forecasting and predictions of destructive weather events enabling decision making that saves lives. Unfortunately, health and social policy have lagged behind in the routine use of these approaches to support policy, planning, monitoring, and evaluation.8

The idea of harnessing systems modelling and simulation to improve decision-making in mental health is not theoretical; it is being achieved across a number of vanguard primary health networks (PHNs). These PHNs are engaging systems modelling experts and a broad range of stakeholders to co-develop these decision support tools, additionally drawing on disparate, multi-agency data sources, best available evidence and the deep understanding and unique perspectives of those with lived experience to inform the modelling. Transparent interactive interfaces of these models (Figure 1) are allowing decision-makers and their stakeholders to critique and alter model assumptions, turn interventions on and off, scale them up and down, stagger their implementation and forecast how different combinations of programs and services will play out in their region over the short and long term to inform commissioning and co-commissioning decisions. As new data comes in, these models are updated to improve their predictive capabilities over time and become a long-term decision support asset. This approach not only provides a more efficient and appropriate predictive planning framework, but its transparency and the democratization of the decision-making process can act to unify communities and align actions to bring about change. In addition, insights from mental health systems modelling applications in Australia have delivered important learning that has facilitated improved regional decision-making.

For example, a systems model developed in partnership with Western Sydney Primary Health Network and their stakeholders9,10 simulated cuts to psychiatric beds under different conditions related to community-based service capacity, forecasting the likely impact on suicide rates over the next 10 years.11 Findings showed that not all reductions to beds result in increases in suicide and that a dynamic ‘tipping point’ exists that is influenced strongly by the availability of community-based mental health services.11 In addition, work carried out as part of the National Suicide Prevention Trial Evaluation funded by the Commonwealth Department of Health, a systems model was developed for the rural population catchment of Western New South Wales (yet unpublished). This work highlighted the likely unintended consequences of implementing general practitioner training (to recognise signs of suicide ideation and refer to appropriate services) together with mental health education programs (aimed at improving mental health literacy and help-seeking). The unexpected increase in self-harm hospitalisations forecast to occur in implementing these two interventions together is explained by a lack of service capacity to meet the increase in service demand these interventions would generate.

Similarly, recommendations made in the Productivity Commission’s report such as the push for comprehensive screening, assessment and early detection of mental health issues, while seeming like a rational, proactive solution, may too produce unintended consequences through the swamping of already stretched service systems, further reducing quality of care. The likelihood of this unintended consequence will depend on regional population and behavioural dynamics, and regional variation in the balance and timing of service or workforce expansion. This balance is one that regional health systems find challenging to establish and maintain given the complexity of the problem and in the absence of the appropriate decision support infrastructure.

In addition, even with perfect implementation, evidence-based interventions can deliver disappointing results. Improving system performance and patient outcomes depend on understanding the critical balance and timing of implementing combinations of interventions and service capacity increases in a given local context. Such insights are impossible through the application of traditional static, linear analytic approaches to decision analysis and evaluation. The task of transforming mental health systems and scaling up effective and contextually relevant strategies to deliver more effective, coordinated, and quality care should not be left to a trial and error approach. There has never been a more important time to leverage the sophisticated decision analytic tools that are emerging in mental health research and practice to provide a blueprint for next-generation investments in mental health.

References

  1. National Mental Health Commission. Contributing lives, thriving communities: report of the national review of mental health programmes and services. Canberra: Available from: http://www.mentalhealthcommission.gov.au/our-reports/our-national-report-cards/2014-contributinglives-review.aspx NMHC, 2014. 
  2. The Senate Select Committee on Mental Health. A national approach to mental health – from crisis to community. Canberra: Australian Government, March, 2006. 
  3. Australian Government Department of Health. Australian Government Response to Contributing Lives, Thriving Communities – Review of Mental Health Programmes and Services. Canberra: Australian Government, 2015. 
  4. Meadows G, Enticott J, Rosenberg S. Three charts on : why rates of mental illness aren’t going down despite higher spending. 2018; June 28: https://theconversation.com/three-charts-on-why-rates-of-mentalillness-arent-going-down-despite-higher-spending-97534. (accessed 21/01/2019). 
  5. Jorm AF. Australia’s ‘Better Access’ scheme: Has it had an impact on population mental health? Aust N Z J Psychiatry 2018;52(11):1057-62. doi: 10.1177/0004867418804066 [published Online First: 2018/10/05] 
  6. Atkinson JA, Page A, Wells R, et al. A modelling tool for policy analysis to support the design of efficient and effective policy responses for complex public health problems. Implement Sci 2015;10:26. doi: 10.1186/s13012-015-0221-5 
  7. Homer JB, Hirsch GB. System dynamics modeling for public health: background and opportunities. Am J Public Health 2006;96(3):452-8. doi: 10.2105/AJPH.2005.062059 
  8. Atkinson JA, Page A, Prodan A, et al. Systems modelling tools to support policy and planning. Lancet 2018;391(10126):1158-59. doi: 10.1016/S0140-6736(18)30302-7 
  9. Atkinson JA, Page A, Heffernan M, et al. The impact of strengthening mental health services to prevent suicidal behaviour. Australian and New Zealand Journal of Psychiatry 2018;https://doi.org/10.1177/0004867418817381 
  10. Page A, Atkinson JA, Campos W, et al. A decision support tool to inform local suicide prevention activity in Greater Western Sydney (Australia). Aust N Z J Psychiatry 2018:4867418767315. doi: 10.1177/0004867418767315 
  11. Atkinson JA, Page A, Skinner A, et al. The impact of reducing psychiatric beds on suicide rates. Frontiers in Psychiatry 2019;https://doi.org/10.3389/fpsyt.2019.00448 
  12. van Os J, Guloksuz S, Vijn TW, et al. The evidence-based group-level symptom-reduction model as the organizing principle for mental health care: time for change? World Psychiatry 2019;18(1):88-96. doi: 10.1002/wps.20609 [published Online First: 2019/01/03] 
  13. Rosenberg S, Salvador-Carulla L. PERSPECTIVES: Accountability for Mental Health: The Australian Experience. J Ment Health Policy Econ 2017;20(1):37-54. [published Online First: 2017/04/19]

Jo-An Atkinson, PhD, is from the Menzies Centre for Health Policy, The University of Sydney and Computer Simulation and Advanced Research Technologies, Sydney, Australia; Ian B. Hickie, MD is from the Brain and Mind Centre, University of Sydney, Sydney, Australia.

About the author
0 Comments
Inline Feedbacks
View all comments
The Mandarin Premium

Insights & analysis that matter to you

Subscribe for only $5 a week

Get Premium Today