Commissioning for health and community sector reform: perspectives on change from Victoria

Additional keywords: commissioning health care, drug and alcohol, mental health.

Commissioning in health and community services broadly encompasses planning, purchasing and monitoring services for a population, subpopulation or individual (Gardner et al. 2016). It is hypothesised that separating the purchaser and the provider of services will lead to efficiencies, reduced conflict of interest and stronger accountability as a result of well-managed contracts (Checkland et al. 2012). However, there are difficulties with achieving these expected outcomes, and commissioning is not a simple answer to solving all design and development issues within systems (Checkland et al. 2012; Addicott 2014).

In Victoria, Australia, governance of health and community services has been devolved for many years. Prior to the recent incorporation of non-clinical mental health community support (MHCS) services into the National Disability Insurance Scheme, these services, along with community-based alcohol and other drug treatment (AOD) services, were commissioned by the State Government. Providers included agencies specialising in the provision of MHCS or AOD services, as well as organisations such as community health services (CHS), which offer comprehensive primary healthcare services. The MHCS and AOD service sectors had evolved over several decades, mostly without a policy framework being developed for the long term (Department of Health 2012). Several reviews and reports had identified issues with the sectors, including access and equity, fragmentation within the system, inconsistency in service models, variable service quality and issues with accountability (Victorian Auditor-General 2011; Department of Health 2012). Many of these problems, which could be seen as arising, in part, from poor commissioning practice, were well known and there was widespread support for reform, including from within the sectors and from peak bodies.

Preparation for reform occurred over several years in the MHCS and AOD sectors and was to have included establishment of consultative and/or advisory mechanisms and the production of a consultation paper, along with developmental projects to inform the technical requirements of key elements in a reformed service system (for a more detailed description of the process in the AOD sector, see Berends and Ritter 2014). In 2013, the Victorian government released frameworks for reform in both sectors, indicating an intent for significant change that would entail redesign of both service systems followed by a process to select a limited number of providers, a process that became known as ‘recommissioning’. The selection processes commenced in August 2013 and operation of the new systems commenced in the third quarter of 2014.

The frameworks for the redesign of each sector specified seven rural and regional and nine metropolitan catchment areas. Each of these would have centralised functions for planning and intake and assessment, which would help integrate services and optimise response to local need. In the AOD sector, the reforms were to focus on adult, non-residential services and were to also include the development of: practice tools and models for improved access to screening, assessment and treatment; analysis and modelling for service planning; development of an activity-based funding model; development of an outcomes-based performance framework; and work to improve information collection, management and reporting (Department of Health 2013a). In the MHCS sector, the focus was on psychiatric disability rehabilitation and support services, and redesign work was to include projects on funding models, demand modelling, client information system functionality, performance management, workforce competencies and piloting of a catchment-based intake and assessment service (Department of Health 2013b). The reform frameworks outlined that recommissioning would entail a competitive two-stage process to select one or more preferred service providers for each catchment. This was to be followed by a 3-month period for transition to the new service delivery arrangements before the date at which implementation would begin.

Overall, the reform was intended to: (1) make the systems stronger, more responsive, streamlined and easier to access and navigate; (2) ensure the needs of clients, carers and families were better met; (3) deliver more flexible funding arrangements for providers; and (4) promote the delivery of consistently high-quality, evidence-based, recovery-oriented services enabling outcomes meaningful to clients (Department of Health 2013a, 2013b). In both sectors, it was intended that the workforces would have high levels of skills and competencies, and accountability would be based on outcomes rather than outputs. There was also to be better integration with clinical treatment services, complementary services funded by the Commonwealth Government and the broader health and human services systems.

This paper presents the findings of a qualitative study on the views of those from the community health, MHCS and AOD sectors in Victoria who were affected by the redesign and implementation of reforms. The aim was to document concerns that emerged about the effect the reforms had on clients and on service delivery at the initial stages of implementation. Feedback was sought on initial support for reform in the system, the reform design process and the transition to the redesigned system. Although the data were collected in 2015, the issues identified at the time remain relevant in today’s policy context and service environment.

Semistructured interviews were conducted by one of the authors (KS) with 23 senior managers or chief executive officers (CEOs) from 12 organisations that provided community health, mental health and drug and alcohol services. The interview schedule was developed by the authors in consultation with colleagues and external stakeholders (Appendix 1). Recruitment was via an email invitation to members of a state-wide network of peak non-governmental organisations, some of which forwarded the invitation to their constituents. Participants were emailed a participant information sheet and consent form and provided consent either by email or verbally at the start of the interview. The interviews were conducted by telephone in February and March 2015, approximately 6 months after the reforms were put in place. Of the interviewees, 13 were engaged in AOD reform, four were engaged in the MHCS process and six were engaged across both processes. Most participants were from agencies that still provided MHCS and/or AOD services in the recommissioned system. Most interviews were conducted with individuals, but in some cases people from the same organisation chose to be interviewed together. Interviews took between 30 and 60 min, based on the depth and extent of the answers provided by the interviewees. All participants had the opportunity to review their interview notes.

Extensive notes were taken during the interviews, and the data were then coded and analysed to identify themes. There was no a priori theoretical framework guiding analysis. Following the general approach suggested by Braun and Clarke (2006), interview notes were read over frequently for familiarisation and initial identification of patterns in responses. Identified themes were reviewed and confirmed through discussions between authors. All participants had the opportunity to review a draft report, and the findings were discussed with the state-wide network before finalising the report.

The study received ethics approval from the La Trobe University Human Ethics Sub-Committee (SHE D15/1).

All participants agreed there had been a need for change in the MHCS and AOD service sectors, and they generally supported the intent of the new policy directions. They had been keen to work with government to develop a system that was more client and family-oriented and had an emphasis on quality, evidence and outcomes:

There had grown up over many years some quirky agencies doing odd things [in the AOD sector], there was no synergy in relation to what happened between regions. There was no sense of standardisation, or a common expectation of a service standard across the State [Manager, Service D].

All they [government] looked at was how many people were getting [MHCS] services. They had no capacity or framework for holding organisations to account [Manager, Service G].

The responsible government department underwent a significant restructure during the redesign period. Informants noted that the subsequent loss of resources and sector-relevant expertise had affected the department’s capacity to undertake the substantial work required for sector-wide reform within the time frames specified. This had contributed to compromises in the processes required for the coproduction and design of key elements of the new service systems, including truncation of consultation and advisory mechanisms, limited use of advice provided by the sectors and failure to complete work to inform the design of systems elements (e.g. central intake functions and reporting systems) before procurement commencing:

Government was grossly under-resourced. You can’t run down the Department and then expect them to manage something like this well…they did the best they could under the circumstances…[the problems] are as much to do with an under-resourced public sector as to do with what they were attempting [CEO, Service I].

In addition, interviewees noted that the quality of services already being delivered had not been formally assessed, nor had existing models of good practice been identified so they could be preserved or built on. Consequently, many informants considered there to be significant flaws in the system design or elements of it. Rural informants indicated that poor processes had resulted in adoption of models designed largely to address issues in metropolitan areas with no flexibility in how they were applied to rural areas.

Concerns expressed by informants about the reform design process were exacerbated by the approach to reform implementation. This included the procurement of elements of the redesigned service system through a highly competitive tender process with strict probity rules and a period for transitioning to the new service provision arrangements. The focus on probity resulted in restricted access to information required for tender development and was particularly difficult in relation to elements of the system where procurement occurred before design work had been finished. This meant some agencies did not have a complete understanding of what would be expected of them when they were developing tenders. Some interviewees attributed the issues that emerged on their inability to ‘do more talking’:

The probity model meant that government stopped talking to the sector and basically required the sector not to talk to each other. In a human services system why would you stop people talking to each other…of course this will end badly [CEO, Service I].

A ‘culture of mistrust’ was seen to arise due to the focus on probity and competition, and, in some instances, well-established collaborative relationships between providers came ‘crumbling down’ because they were now competing to offer services. Some community health informants argued that their agencies, which had done significant work to establish comprehensive primary care with integrated MHCS and AOD services, were not seen as adding value to the service system by reducing fragmentation but were simply viewed as another competing agency:

It is clear to me that [they]…looked at every tender in isolation from the system – they looked at someone’s tender and said, ‘can they deliver this’? They didn’t look to see what would make sense for that area…In some areas they have new providers coming in and the old providers are still there doing their old work with different funding streams – there are ludicrous outcomes. It felt like it was driven by a procurement strategy rather than what outcomes they were trying to achieve [Manager, Service G].

The financial burden to agencies of developing proposals and participating in the procurement process was high, as was the human cost from uncertainty about future funding and employment for staff. Uncertainties about future funding meant that many organisations were not able to offer staff continuity of employment, and effects on staff well-being and morale were significant:

It’s very costly – the amount of existing resource and time that gets caught up and lost in going through a process like this is significant – not just for management staff. There is a high level of concern from all staff about both the models they hold dear to their hearts and their jobs. The cost to the client group with respect to loss of stability is also very high [CEO, Service D].

It was a horrible, stressful time. People felt terrible. It made the environment bad [Manager, Service B].

After the successful agencies were announced there was a 3- or 4-month transition period from one agency to another. This was reported to be highly disruptive for both successful and unsuccessful tenderers. Informants maintained that there were too many things happening at once and not enough time to organise properly as they closed defunded services and/or established new services, established new models of care, transitioned clients and dealt with significant workforce and human resources issues. This was compounded by reported deficits in the information, specifications, guidelines and tools to assist with establishing the new system.

Some interviewees reported losing funding in areas where they already operated and gaining funding in regions where they did not have services. This led to workers being made redundant and offices closing in one area while the organisation was finding new accommodation and employing staff in another. In addition, if an agency was defunded the staff could not transition to a different agency, so redundancies needed to be paid and continuity of employment was lost. Agencies that had lost funding were expected to continue providing services until the new arrangements became operational, so they needed to manage staff who would soon be redundant, as well as clients who were upset about having to change service provider. It was noted that the process had resulted in a very expensive ‘massive restructuring of the labour force’:

Because it was a different model, AOD staff had to be made redundant [from our agency] and then offered a job in the new program [in another agency]. Half of our staff had to be paid redundancy – which is a gross waste of public funds. This had to come from funds that should have been used for programs – we lost knowledge, skills and relationships in the sector. Some of the staff didn’t even apply for jobs with the other providers because they were so disillusioned – instead they left the sector altogether [CEO, Service H].

Agencies also reported lack of concern about the realities of their business by government, with the transition process beginning before the service contracts became available:

We took on a lot of risk in this process. We didn’t see the contract until three months after it commenced. We were making staff redundant, taking on new staff, taking leases on buildings. There was a lack of understanding in the Department of the business realities of [the agency]. Even for the successful organisations the transition period was challenging – I doubt that the Department realises how much good will or taking on of risk by new providers was required to get the thing established. It really was the good will of the sector, including some of the defunded organisations that got it across the line [CEO, Service K].

MHCS and AOD service providers indicated that that there did not appear to be a well-planned process for transitioning clients between defunded and newly funded services. This meant that some newly funded agencies were still in the process of establishing their service and recruiting staff when defunded agencies were trying to hand over clients. Compounding this, it seemed the model for the new systems had not been widely communicated to clients and other providers (e.g. GPs) so that many did not know that access would be managed through a central intake service:

There was an expectation that the agencies would do the change management in the community – this really should have been [DHS] responsibility. We still have clients and GPs who don’t know how to get people into the system [Manager, Service L].

Overall, there was significant demoralisation among informants, who were disappointed with the process of designing and implementing the reforms, although there were some interviewees who were optimistic that, over time, the changes would produce improvements for clients, despite the initial difficulties.

Although not convinced that the method of producing the reform was the best one, 6 months after implementation some informants considered that, despite the issues, the new arrangements could result in a more coherent, easier to navigate system with reduced variance in practice, a better interface between community support and clinical services and improved capacity to understand demand and allocate resources accordingly.

However, most informants reported significant system deficits that they attributed to the recommissioning process. These largely pertained to loss of some of the perceived strengths of the pre-existing systems that occurred as agencies were defunded or disrupted by elements of the new approach. For example, some noted that the centralised intake and assessment processes had a negative effect on a range of collaborative models for improving service coordination, such as direct referral (e.g. between acute and primary care sectors), integrated and comprehensive primary health care provided by one organisation (e.g. CHS that had previously conducted comprehensive assessments and coordinated service delivery) and innovative models to reach specific groups, such as those experiencing homelessness, those interacting with the justice system or those with dual diagnoses:

All this [work in partnership] has fallen away. We have spent years trying to develop referral pathways. The new system is crude and has utterly misunderstood how this [referrals between providers] happens in the first place [CEO, Service I].

In addition, there was also a set of concerns related to implementation of the centralised intake systems, which were reported to create barriers for clients, and there was a loss of clients from the system, although the reason for this was not clear.

There was also emerging scepticism about the intent of the reforms, with some informants considering that rather than working towards an integrated system delivering high-quality services at a reasonable cost, simply reducing the number of service providers became a prime objective. They considered development of a quality and outcomes framework that could have then been used to guide decision making about which agencies were to be funded may have been a better approach. This kind of speculation led some to believe they had been misled and that there had not been enough transparency about what the reforms were aiming to achieve.

Reforming sectors is known to be difficult, disruptive and to consume a great deal of time and resources. Therefore, the intended outcomes need to be worth the costs, both human and financial, involved. In complex systems, like the sectors discussed here, there has been recognition that more effective solutions can often be generated by moving towards collaborative processes in which multiple stakeholders with a range of expertise contribute to design, rather than a ‘command and control’ approach (World Health Organization 2008). Had a collaborative approach been adopted, as initially planned, some of the issues emerging may have been avoided. Using those with expertise in designing reform in addition to conducting some pilot projects in critical areas may have led to the adoption of a more tailored system and the prereform work would have been completed before implementation. Ongoing monitoring and evaluation of the new system would still be needed to ensure the implemented changes are on track to produce the desired results and to revise systems if necessary.

It is difficult to determine which part of the process of reform was most responsible for the problems that emerged because interviewees reported difficulties with procurement, sector redesign and ultimately implementation. Ultimately even among those who were delivering parts of the new system, many were disillusioned. If there are further attempts at reform in the future, it may be difficult to gain the cooperation and engagement of providers.

A putative strength of commissioners is their separation from the service provision function, which theoretically enables them to consider the best interests of service users and design systems accordingly. One of the key challenges in realising this advantage is having the resources and sophisticated skills within the commissioning workforce to plan and design effective and efficient service systems followed by contracting. The danger in not investing in rigorous planning and design is a propensity to revert to a reliance on competitive tendering processes to address systems problems. On the financial side, it appears that poor commissioning practice effectively resulted in resources being used in the wrong parts of the system; that is, insufficient investment in redesign and planning resulted in higher than necessary resource utilisation by service providers as they competed for funding and implemented reforms. A further effect was loss of the widespread enthusiasm for reform, as well as some of the benefits that may have accrued from using sector expertise.

Models based on regional structures have an advantage in that they can enable jurisdiction-wide initiatives to be effectively adapted so that implementation works well locally. A regionalised, sector-wide model using central intake and assessment and a smaller number of service providers (or consortia) can have advantages, including improved ways for clients to determine where to go to seek help and better allocation of resources to clients.

However, in this instance the procurement process seems to have undermined some of the strengths of the existing services, and may not have led to the ‘best buys’ for local conditions, particularly in rural areas. The lesson here appears to be that in procuring elements of human service systems it is important to treat the system like a system (rather than simply viewing each organisation as an individual competitor) and consider existing strengths, capacity and relationships rather than undermining them. This lesson would be in line with the recommendations of the Shergold report, which identifies the importance of strong collaboration for effective delivery of human services (Shergold 2013).

Sector reform is a difficult process in a service delivery system that has become large and fragmented. Although there are potential benefits in recommissioning entire service sectors, there must be significant investment of time and resources in sector planning, design and development before implementation, as well as excellent planning and communication with the relevant organisations. Disenfranchising service providers who want to see reform can limit the significant benefit to be gained from their knowledge, expertise and enthusiasm, which, in turn, potentially reduces the benefits good system design can bring to consumers, carers and communities.

Virginia Lewis is Editor-in-Chief of the Australian Journal of Primary Health. Kate Silburn was a Co-guest Editor of the Journal’s 2016 special edition, ‘Is Commissioning the Solution for Better Health Outcomes?’. Despite this relationship, they did not at any stage have editor-level access to this manuscript while in peer review, as is the standard practice when handling manuscripts submitted by an editor to this journal. Australian Journal of Primary Health encourages its editors to publish in the journal and they are kept totally separate from the decision-making process for their manuscripts. The authors have no further conflicts of interest to declare.