Latest evidence casts further doubt on the effectiveness of headspace

Steve Kisely and Jeffrey CL Looi
Med J Aust 2022; 217 (8): 388-390. || doi: 10.5694/mja2.51700
Published online: 12 September 2022

Despite significant government expenditure on headspace, youth mental health outcomes still need improvement

Headspace has formed a major part of the federal government’s approach to mental health care for young people since 2006. Its clinical services can be delivered in person or remotely through telepsychiatry, online and telephone support; headspace also offers vocational guidance and education programs for schools. Funding over two decades has been substantial. This includes $35 million in 2017–18 and a further $263 million over the following 7 years, as well as $109.7 million for early psychosis interventions, which are run, in part, through headspace.1 Concerns over the mental health consequences of the COVID‐19 pandemic have led to even more funding. Of the $2.3 billion allocated in the 2020–21 federal budget on new mental health expenditure, $765.8 million was directed to headspace, as well as related services for older age groups in Head to Health centres.2 This was to reduce wait times, continue or upgrade five existing services, and establish ten new centres with the aim of reaching 164 centres nationwide by 2025.3 A further initiative was the creation of the National Student and Graduate Placement Program to bolster the mental health workforce.3

In the following year, the 2021–22 federal budget allocated an additional $14.3 million in funding for headspace, including schools‐based suicide prevention and digital work and study programs.4 This did not include funding through either Medicare or state governments for clinicians based in over 100 centres, as well as $45.7 million over 3 years to extend the vocational support program from 24 to 50 sites.1,3

As a result of this financial support, headspace has a significant presence in the delivery of mental health services to young people. For instance, there were 441 914 care episodes in 2020–21 and, since inception, headspace has provided 4.4 million services to more than 700 000 young Australians.3

This expansion has occurred despite limited evidence of effectiveness, with many studies using either process measures or uncontrolled satisfaction surveys. Where there are data on outcomes using standardised instruments, findings have been disappointing. A 2009 report found that for 2222 participants, scores on the Kessler Psychological Distress Scale (K10) decreased from a mean of 28.1 to 26.9 in the 40% who completed the survey.5 A 2015 study, with a highly unrepresentative follow‐up rate of 3.1%, found that 78.9% of the sample experienced no clinically significant benefit on the K10,6 and only 37% showed clinically significant psychosocial improvement on the Social and Occupational Functional Assessment Scale. In 2015, another study reported an improvement in mean K10 scores of only 2.3 points, with just 13.3% reporting a clinically significant improvement.7 A small uncontrolled study from September 2012 to July 2017 (n = 77) did report statistically significant benefits on the K10.8 However, the proportion of participants in a lower clinical stage of illness at the end of treatment was not statistically significant.8 These findings are reflected in a more recent study of 1510 young people that found only 35% had good functional outcomes after 24 months in the period 2008–2018.9 The available evidence is further limited by the relatively small sample sizes, the highest of which was 2222 participants from 2008–09, less than 0.5% of the 700 000 young people seen by headspace since 2006.3 This could have meant that studies were either underpowered to detect an effect, or that the findings were unrepresentative and lacked generalisability. Depending on the direction of bias this might have resulted in effects being over‐ or underestimated. As a result, the ongoing level of uncertainty in the strength of the available evidence seems inconsistent with the current generous funding of headspace.

Concerns about the effectiveness of headspace and its associated programs are not new. The Productivity Commission highlighted the limited benefits and recommended ending a requirement that Primary Health Networks fund headspace.10 Instead, the youth service should have to compete for funding, thereby allowing Primary Health Networks to redirect funds to services that best meet their population’s needs. Others have noted that despite the significant expenditure on headspace, youth mental health outcomes have worsened over the same period in terms of population‐based psychological distress, self‐harm and suicide.11

Another concern has been the lack of integration with other services. For instance, in 2014, the National Mental Health Commission noted that the creation of headspace mental health centres was conducted without sufficient consultation, leading to “duplication of, and competition with, other community, private and state government services”.12 Australia’s mental health system is already complex and fragmented, with roles and responsibilities split between federal and state, public and private, and community and acute services. Large non‐governmental organisations such as headspace further complicate this picture, increasing fragmentation in an already poorly integrated national mental health system. This level of complexity is not seen in any other branch of medicine.

An evidence‐based response to these concerns might be a re‐evaluation of the model and redirection of funds elsewhere, such as primary care or under‐resourced state‐provided mental health services. Instead, there have been calls for even more funding directed at expanding the scope of headspace and related services towards the “missing middle” — moderately unwell people who do not require crisis services but still need ongoing treatment, which is often beyond their financial means.13 This is despite the fact that conventional headspace centres may be ill‐equipped to manage this population as just under 40% of clients have no recorded diagnosis.14 Given this limitation, the suggested model for the expansion to the missing middle is therefore based on existing early episode psychosis services, already linked to headspace and said to have a strong evidence base.13 However, in comparison with treatment as usual, the advantages of these services are restricted to the initial 2 years with limited evidence beyond that.15 There are also concerns about value for money. In an economic evaluation, the incremental cost effectiveness ratio for these services was $318 954, which exceeded the $213 000 threshold set by the Cabinet Office of Best Practice Regulation.16 Importantly, early episode psychosis services are not targeted at people with other diagnoses such as mood, personality or substance use disorders, who may have very different needs from those with psychosis.

Australia’s mental health system does not need further complexity but rather should return to evidence‐based care delivered in public and private settings. Most information on effectiveness concerns inpatient settings and data from the Australian Institute of Health and Welfare and the Private Psychiatric Hospitals Data Reporting and Analysis Service suggest high effect sizes that easily exceed those of headspace.17 Ideally, funding for headspace would therefore be better spent to increase the accessibility and affordability of existing services, as well as to address social determinants including early life experiences and workplace mental health.10 At a minimum, headspace services should be situated and funded within state/territory publicly provided mental health services with established systems for clinical governance and subject to a similar level of accountability. This includes the same key performance indicators that are mandated for public mental health services and available through the Australian Institute of Health and Welfare, as well as care pathways of referrals, interventions, and the discipline of any onward referral. In particular, the routine use of standardised clinician and patient reported outcome measures, such as the Health of the Nation Outcome Scales,18 would enable comparisons to existing public and private sector services. In terms of links to the wider mental health system, some headspace centres in Melbourne are already integrated with public mental health services, allowing a single point of entry to coordinated services.19

In addition to the integration of existing centres into mainstream mental health services, alternatives to headspace include the diversion of proposed future funding to public sector mental health services where there is better evidence for improved outcomes, or to private psychological and psychiatric services through the Better Access program.20 For instance, the Australian Mental Health Outcomes and Classification Network reports outcome data for public mental health services. In 2019–2020, baseline psychiatric symptoms for new outpatient referrals as measured by the Health of the Nation Outcome Scales were 11.1 (standard deviation, 6.2) for 15–24‐year‐olds compared with 7.4 (standard deviation, 5.8) at discharge to no further care.21 This approximates an effect size of 0.62, which represents a change of medium magnitude in terms of Cohen's classification (small, 0.2; medium, 0.5; large, 0.8).22 Although there are no outcome data for private outpatient psychiatric services, an evaluation of focused psychological therapies under the Better Access initiative is currently underway, and this will include data for child and youth patients.23

It is time for a national discussion about youth mental health in the context of declining mental health outcomes in young people, rather than doing the same thing repeatedly and expecting different results. A single funding stream linked to uniform reporting requirements would help to reduce the existing duplication, confusion and fragmentation in care for this population.

Provenance: Not commissioned; externally peer reviewed.

  • Steve Kisely1
  • Jeffrey CL Looi2

  • 1 University of Queensland, Brisbane, QLD
  • 2 Australian National University, Canberra, ACT


Open access

Open access publishing facilitated by The University of Queensland, as part of the Wiley ‐ The University of Queensland agreement via the Council of Australian University Librarians.

Competing interests:

No relevant disclosures.


remove_circle_outline Delete Author
add_circle_outline Add Author

Do you have any competing interests to declare? *

I/we agree to assign copyright to the Medical Journal of Australia and agree to the Conditions of publication *
I/we agree to the Terms of use of the Medical Journal of Australia *
Email me when people comment on this article

Online responses are no longer available. Please refer to our instructions for authors page for more information.