Evidence or ideology? Let the Debate Continue … More evaluation please!

Evidence or ideology? This was the question posed by Allison et al. (2015) regarding the basis for the National Mental Health Commission’s recommendation to transfer AUD1 billion from acute to community sectors, a position eschewed by Minister Ley upon the leaking of the Commission’s Report (Ley, 2015). Allison and colleagues concluded that further psychiatric bed closures were not a defensible option, implicitly suggesting that the Commission’s primary driver was ideological. The position of Allison and colleagues was based on

• The level and breadth of evidence utilised in the determination of the Commission’s recommendations, the Commission’s position being based on KPMG modelling of 7 ‘optimal care’ scenarios heavily reliant on expert opinion;

• Low mental health bed numbers in Australia relative to other Organisation for Economic Co-operation and Development (OECD) countries;

• The example of the impact of acute bed closures on Emergency Department (ED) presentations in South Australia.

The Commission’s acceptance of KPMG’s modelling, modelling predicated on ‘up-front costs to increase access to services across all severity levels’ (KPMG, 2014), is, we agree, fraught with concern. Not only did KPMG ‘advise the Commission to be wary of using data inappropriately’ (Henderson Editorial on Allison), KPMG’s modelled savings for ‘optimal care’ are to be realised over a 9-year time frame subsequent to initial investment. Thus, the KPMG modelling rests on an ‘invest to save’ framework rather than the Commission’s ‘reinvest to save’ position.

The fact that bed closures have funded community care can be affirmed not only at a system level but also at the individual patient level. In a comparison of the average cost profiles for individuals currently treated in public specialised mental health services (inpatient, outpatient, ambulatory and community services) between the First and Second National Surveys of Psychosis (Neil et al., 2014), we found a marked shift in costs from inpatient care to community care through both public specialised and non-governmental organisation (NGO) providers, a shift consistent with the objectives of the Third and Fourth National Mental Health Plans. However, alongside this shift was an increase in mental health–related ED presentations, these being the primary indicator that concerned Allison and colleagues. Excluding presentations not referred for admission (to ensure consistency across the two surveys), average per person costs for mental health–related ED presentations tripled in the intervening decade. The increase in costs was driven by a 2.8-fold increase in the number of patients seeking care in the ED, and by a 4.9-fold increase in visits per person. We concur with Allison and colleagues that this increase is a sign of a system in distress. Furthermore, we found no change in the extent of productivity losses for individuals living with psychotic illness, and the proportion receiving disability support payments increased from 68.3% to 73.7%. For perspective, a reduction in Disability Support Payments is the primary source of savings being relied upon by the Commission. Furthermore, we observed an increase in productivity losses for carers, associated with an increase in the number of carers receiving the Carer Payment, and the time they spent caregiving. Moreover, despite over a halving of mental health inpatient costs for people living with a psychotic illness between the First and Second Surveys, there was no commensurate increase in costs for supported accommodation.

In the absence of an improvement in productivity, the shift in resources towards community service provision and the community sector raises a number of questions. First, has the shift been associated with an improvement in other outcomes necessary for increased effectiveness and efficiency of care provision? For example, has level of function and/or quality of life of individuals with severe mental illness improved? Second, are people with severe mental illness and their families and carers more exposed to the burden of daily care provision for psychosis? If so, from their perspectives, has the shift in burden been worthwhile? Such valuations will underpin the sustainability of daily care provision by carers and families. Are more resources justified to support the shift towards community care provision, and if so, where should they go?

The answers to these questions remain unclear. This leads us to concur with other recommendations of the Commission pertaining to the necessity for increased research and evaluation. It is only through research and evaluation that we will truly be able to ascertain those interventions and system changes that are effective and efficient, through which long-term benefits and savings will likely be realised. Such information is imperative to robust policy development and implementation.

‘[G]oodwill alone cannot make up for restrictions imposed by limited funding of services’ (Mental Health Services Commission, 1982). Like Allison and colleagues, we would argue that if mental health is really a priority, additional funding should be allocated, even if only with the objective of reducing the need to provide Disability Support Pensions for people with mental illness and thus minimise costs in the longer term.

See Debate by Allison et al., 49(11): 960–962.