Programme budgeting is a technique designed to identify how much money has been invested in major health programmes (Reference Brambleby, Jackson and Muir GrayBrambleby et al, 2007). Information concerning levels of investment in mental health and other healthcare areas can be found on the Department of Health website (www.dh.gov.uk/en/Managingyourorganisation/Financeandplanning/Programmebudgeting/index.htm). Marginal analysis is an economic appraisal that evaluates incremental changes in costs and benefits when resources within a programme are increased, decreased or deployed in different ways (Reference Brambleby, Jackson and Muir GrayBrambleby et al, 2007).

The practical application of programme budgeting and marginal analysis (PBMA) has been described by Ruta et al (Reference Ruta, Mitton and Bate2005). The approach can be broken down into five essential steps. Steps 1 and 2 establish the total resources available and identify services on which these resources are currently spent. They allow for the relevant programme budget to be calculated. Step 3 involves identifying potential services as candidates for receiving additional or new resources (and the costs of potential benefits of putting such resources into growth areas) – the ‘wish list’. An assessment of existing services is then undertaken (step 4) to establish which can be provided equally effectively and with fewer resources, thus releasing resources to fund items on the wish list. Step 5 is arrived at when some items on the wish list still cannot be funded. It identifies services that should receive fewer resources, or even be stopped, because greater benefit would be reached by funding other items on the wish list. This collection of services becomes the ‘hit list’.

Fundamental to the above process is an economic evaluation of the marginal benefit gained per extra unit of resource for one service and the benefit lost from having one unit less in another. It is usual to form an advisory panel which will examine the costs and benefits of the proposed changes and make recommendations which can be acted upon by commissioners. The establishment of the advisory panel provides an opportunity to include a wide range of stakeholders, including service users and carers, as well as managers and clinicians.

Method

The methods for conducting a PBMA analysis discussed in Ruta et al (Reference Ruta, Mitton and Bate2005) were adapted for the local pilot. A marginal analysis advisory group was formed to discuss possible changes to mental health services and to examine their costs and benefits. Five meetings were organised between April and October 2007. The membership was selected from locally established networks and included professional and managerial staff from the main NHS provider (NWMHP), commissioners, a public health consultant, representatives from the independent and charitable sectors, and representatives of service users and carers. General practitioner (GP) representation was also sought, although no GP ever attended any of the meetings.

The programme budget under consideration was Norfolk Primary Care Trust's NHS expenditure on mental health for the fiscal year 2006/7. It was assumed that no additional funds would be made available for the following year so that any judgements about investment would have to take into consideration the Trust's disinvestment in mental health. It was acknowledged that considerable disinvestment in mental health had already occurred and that the Trust planned a further disinvestment of £2 million, specifically from older peoples' services in the fiscal year 2008/9.

The advisory group was responsible for: developing and weighting a set of criteria that services could be judged against; identifying options for service change (the wish and hit lists); developing the possible options for change into well-defined outline business cases; and, finally, scoring these cases against the criteria.

All of the business cases originated with, and were subsequently developed by, members of the advisory group. A standard business case template was used and completed based on information from the professional literature, data held by local agencies and stakeholders, and from local professional opinion.

The average scores for each criterion were multiplied by the corresponding weights and summed together to produce an overall weighted benefit score. The business cases were then prioritised in terms of cost per person per benefit point (cost:value ratio; Reference Wilson, Rees and FordhamWilson et al, 2006).

Results

The advisory group identified 23 criteria against which to assess mental health services. These were clustered into four groups covering effectiveness, quality, policy and strategy, and feasibility. Effectiveness and quality were weighted 30% each, whereas policy and strategy and feasibility 20% each (Fig. 1).

A number of potential services for investment were identified. These included:

1. • mental health promotion in the community (e.g. reducing stigma);

2. • complementary interventions (arts, nutrition, exercise, acupuncture, social activities, relaxation techniques);

3. • clinical treatments (specifically, the provision of community-based psychological therapies, a nurse-led eating disorder service and an expansion of the existing assertive outreach services for severe and enduring mental illness);

4. • lifestyle support for service users and carers within the community;

5. • young peoples' services (such as a ‘one-stop shop’ for people with emotional problems aged 13–25 years).

Fig. 1. Criteria and weightings identified by the advisory group to assess mental health services.

The advisory group also identified potential items that could receive fewer resources. These included:

1. • reducing the volume of antidepressant prescribing in primary care;

2. • enforcement of a protocol for prescribing atypical antipsychotic medications devised between primary and secondary care;

3. • out-of-area specialist treatment placements;

4. • developing a newer model of delivery for crisis resolution and home treatment services;

5. • rationalising family support teams;

6. • restructuring day services, particularly within older peoples' provision;

7. • alcohol and drugs services.

There was the potential to release £3.77 million to fund the service expansion options. Business cases were then developed for six specific service areas.

1. 1. ‘Gardening for health’ project.

2. 2. Young persons' one-stop shop.

3. 3. Assertive outreach services.

4. 4. ‘Floating support’ (outreach tenancy support workers and benefit specialists).

5. 5. Holistic mental well-being service (cognitive–behavioural therapy, support in the workplace, complementary therapies and guidance for self-management of well-being).

6. 6. Nurse-led eating disorders service.

The costs per patient and benefit (out of a maximum score of ten) for each business case are shown in Fig. 2. Interestingly, expansion of the assertive outreach service was associated with the greatest additional benefit; however, this option was also the most costly per patient. In comparison, the marginal benefit of a holistic well-being service was judged to be almost as beneficial but significantly less costly.

The advisory group calculated that £194 000 would be released through changes to prescribing practices; this sum should be used to fund a holistic mental well-being service and to develop a young persons' one-stop shop. Additional sensitivity and threshold analyses added further weight to this conclusion.

Conclusions

The central importance of commissioning healthcare to achieve high-quality and personalised services that improve health and well-being within the population is a core plank of Department of Health policy (Department of Health, 2007). Decisions taken by commissioning bodies are likely to have a profound effect on the shape of mental health services at the provider level within England during the coming 5 years. However, the resources that are going to be available to commissioners are likely to be limited, as the rate of spending increase on the NHS observed for the period 2002-2007 is going to be considerably curtailed for the period up to 2011 (Reference Wanless, Appleby and HarrisonWanless et al, 2007). It is essential, therefore, that commissioning decisions are taken within an objective framework and in as inclusive a way as possible.

The future commissioning of mental health services is likely to arouse much controversy. Historically, the provision of mental health services has represented a broad church containing a wide spectrum of stakeholders and other interested parties. All stakeholders have a view about service provision and want this view heard. Although not without its limitations, PBMA provides one such framework that allows stakeholders to feel included within the decision-making process. It also provides an element of transparency concerning how decisions are taken. Ultimately, responsibility for commissioning decisions must lie with primary care trusts and their boards, but PBMA provides a framework that can inform a primary care trust board.

Fig. 2. Cost and benefit calculated for each business case identified by advisory group.

This pilot demonstrated that it was possible to engage (with the exception of GPs) a wide range of stakeholders who were prepared to work together and collectively provide recommendations to a primary care trust about investment and disinvestment decisions in mental health. Furthermore, the PBMA model was shown to provide a useful framework that enabled a structured approach to be applied to the decision-making process. The stakeholder members of the advisory group were generally satisfied with the outcome, and believed that their views had been incorporated into the process of decision-making.

Declaration of interest

None.