- Email: [email protected]
Maximising health gain within available resources in the New Zealand public health system
Health Policy 55 (2001) 37 – 50 www.elsevier.com/locate/healthpol
Maximising health gain within available resources in the New Zealand public health system Paul Bohmer a,*, Charles Pain b, Alex Watt c, Paul Abernethy a, Janet Sceats d a
Southern Regional Health Authority, P.O. Box 3877, Christchurch, New Zealand b Midland Regional Health Authority, P.O. Box 1031, Hamilton, New Zealand c Southern Regional Health Authority, P.O. Box 5849, Dunedin, New Zealand d Midland Regional Health Authority, P.O. Box 1031, Hamilton, New Zealand Received 4 March 2000; accepted 3 July 2000
Abstract Objecti6e: to obtain intervention-condition-specific investment and disinvestment recommendations which optimise the potential for health gain from existing respiratory diseases resource expenditure; and to trial a health economics technique for this purpose, assessing its usefulness as a means of prioritising health services resource allocation. Design: a programme budgeting and marginal analysis (PBMA) exercise drawing upon the expertise of an advisory group of clinicians, managers and consumer advocates, supported by health authority staff. Setting: the Southern and Midland health regions in New Zealand which have populations of one million and seven hundred thousand respectively. Health system context: publicly funded secondary care sector in which regional health authority (RHA) purchasers contract for services with health care providers. Methods: available evidence on the marginal costs and benefits of services for respiratory diseases was examined by an advisory group who produced investment and disinvestment recommendations by consensus using agreed prioritisation criteria. Results: A list of specific investment and disinvestment proposals. Implementation plans for a number of investments formed part of the business plans for both RHAs in 1997/1998. No disinvestments were planned. Conclusions: prioritisation methods like -PBMA, which are explicit and rational, can produce defensible evidence-based recommenda-
* Corresponding author. Present address: Health Funding Authority, Private Bag 92-522, Wellesley Street, Auckland, New Zealand. Tel.: +64-9-5809057; fax: +64-9-5809001. E-mail address: [email protected] (P. Bohmer). 0168-8510/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 5 1 0 ( 0 0 ) 0 0 1 0 7 - X
38
P. Bohmer et al. / Health Policy 55 (2001) 37–50
tions with the additional benefit of the credibility and support of an expert advisory group. The process encourages co-operative working and may itself have enduring benefits. However, preparation and conduct of such exercises is resource intensive and requires careful planning. This exercise has provided valuable lessons for the conduct of future prioritisation work. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: PBMA; Prioritisation; Secondary care; Respiratory services; Purchasers; New Zealand
1. Introduction This paper describes a prioritisation exercise conducted by two purchasers in the New Zealand Health System. This has provided a unique opportunity to compare the outcome of two such exercises in two different localities but in the same health service programme area. The methods and results are briefly described but the main emphasis is on discussion of the practical lessons learned, which may be valuable in future applications of this method.
1.1. Health system context In New Zealand, as in other health systems, growing demand for services has placed increasing pressure on health care resources. This has led to an emphasis on obtaining ‘value for money’ and an awareness of the opportunity costs of spending scarce resources on marginally effective or even harmful interventions. The regional health authorities (RHAs) were established with the implementation of the New Zealand Health Reforms in 1993 and were funded by means of a population-based funding formula (PBFF) to purchase public health (health promotion and prevention), personal health (family health and hospital services) and disability support services for the residents of their respective regions. In 1996, most New Zealand health services were purchased on a regional basis by one of four RHAs. These RHAs were responsible for improving, maintaining and protecting health [1]. Their role was to take an overview of the population’s need for health services and to optimise the effectiveness, efficiency, equity, acceptability, safety and risk management of service provision. A key part of this role was to prioritise services and interventions in order to maximise health gain for the populations served. However, much of the purchasing function has been carried out on a ‘role-over’ basis, with minimal change to the pattern of service provision. A means of making rational prioritisation decisions within an explicit cost-benefit framework is needed to enable the purchasers to make decisions which optimise health gain within the existing resource constraints. An economics-based model such as programme budgeting and marginal analysis (PBMA) offers a possible method for achieving this. And because PBMA takes into account needs, costs and benefits, it offers an advantage over needs assessment alone by considering outcomes and costs of treatment services in making prioritisation decisions.
P. Bohmer et al. / Health Policy 55 (2001) 37–50
39
1.2. PBMA PBMA is a method, which has been used in New Zealand and in other health systems, which involves evaluation of the costs and benefits of existing and potential services or interventions [2–5]. It comprises two activities: 1. programme budgeting, examining evidence on how the current resources of a defined programme are allocated; and 2. marginal analysis, which involves an evidence-based evaluation of the service mix to determine whether greater benefits could be realised by changing the proportionate allocations of resources among the various services or interventions. This involves making value judgements, supported by evidence, of whether greater benefits for the population could be achieved with the same resources by disinvesting in less effective services and reinvesting in more effective services. A framework for decision-making is thus provided in which value judgements are explicit and informed by evidence. An advisory group, comprised of individuals with relevant expertise and experience in the programme area and others who represent various interest groups, can be employed to make prioritisation decisions using this framework while applying their own value-judgements. For more details on PBMA, the reader is referred to Health Policy, 35 (2) (1995), which is entirely dedicated to PBMA.
1.3. Project background The Midland and Southern regional health authorities (MRHA and SRHA) decided to implement a project to assess the feasibility of applying PBMA as a means of prioritising resource allocation within an existing budget, and to obtain specific recommendations regarding the optimal allocation of resources in the important programme area of respiratory diseases. Respiratory diseases consume a significant proportion of health care resources. For example, in the Midland Health region, respiratory diseases consume approximately 7% of the total expenditure. The burden of these diseases is considerable. They account for some 7.2% of hospital admissions per year in the Southern region and 9% of admissions in the Midland region. They also account for some 13% of all the deaths in the Southern region and 16% of deaths in the Midland region. From the RHA perspective, therefore, they provided an opportunity to employ PBMA in an important area with the possibility of increasing health gain for patients within existing resources. 2. Methods The procedure for applying the PBMA methodology to the respiratory programme was developed jointly by representatives from the SRHA and MRHA, and involved an advisory group taking part in two workshops. Fig. 1 illustrates the PBMA procedure employed.
40
P. Bohmer et al. / Health Policy 55 (2001) 37–50
Fig. 1. PBMA process.
2.1. Phase I International classification of disease (ICD-9) codes were used to define the diseases included in the programme. This approach defined the respiratory programme in a way which ‘cut across’ the currently purchased respiratory and other programmes, for both of the RHAs. For example, lung cancer is currently included in the ‘oncology programme’ and not the ‘respiratory programme’ of both the Authorities. This broad condition-based approach to programme definition was adopted by both RHAs for consistency and because it is in keeping with the logic of an integrated disease management approach [6]. The conditions included in the respiratory programme for this exercise are given in Table 1. An inclusive programme budget was derived for all these conditions, comprising health promotion and prevention, primary health care, outpatients and secondary care, and laboratory and pharmaceutical costs. There was some difficulty in establishing these budgets as much of the available data was inadequate, particularly in the primary care sector. However, costs were based on best estimates and reflected the purchase costs of the services to the health authorities. Though they did not necessarily reflect actual hospital costs, they were likely to have been reasonable estimates. Other important activities in Phase I included:
P. Bohmer et al. / Health Policy 55 (2001) 37–50
41
Table 1 The respiratory programme as defined by ICD-9 codes Diseases
ICD 9 Classification
Tuberculosis Lung and pleural cancer Cystic fibrosis Acute bronchitis and bronchiolitis Pneumonia and Influenza COPD and allied conditions Asthma Pneumoconiosis Other diseases of the respiratory system Sleep apnoea and other sleep disturbances
11 162 277 466 480–486 491, 492, 494, 496 493 501.506, 507 510–518 780.5
1. identifying and inviting an Advisory Group with members chosen to provide representation from across each region and relevant expertise in the various health care disciplines involved in respiratory diseases management and care including respiratory physicians, ENT surgeons, general practitioners, oncologists, nursing staff, physiotherapists, hospital managers, academics, consumer groups, lay community members and representatives for Maori (the Advisory Groups in both the RHAs comprised of approximately one-third respiratory physicians and the majority of members in both groups were doctors); 2. collection of background data to provide an overview of disease burden and costs related to the respiratory programme; and 3. an Advisory Group workshop to produce a detailed description of the existing respiratory disease programme, discuss and identify potential candidates for investment or disinvestment and agree benefit criteria to be used in the marginal analysis. At the first workshop a disease and intervention framework was used to identify and list the full range of existing and potential interventions for each stage of all diseases in the respiratory programme. This framework was developed for the exercise and is shown in Table 2. Interventions were grouped under the headings of prevention, assessment and investigation, treatment, continuing care, support and rehabilitation. Table 2 Intervention/disease framework Type of intervention Tuberculosis Lung Cystic Pneumonia Acute COPD Etcetra cancer fibrosis and bronchitis and influenza bronchiolitis Prevention Assessment and investigation Treatment Continuing care, support, rehabilitation
42
P. Bohmer et al. / Health Policy 55 (2001) 37–50
Cost data were obtained for the existing respiratory programmes in each RHA. These data were collected from a number of sources. For public health an estimate was made of the costs of health education, smoking cessation programmes and health education for asthma. In primary care these costs consisted of both RHA subsidies and private fees paid for a GP consultation, pharmaceutical subsidy costs and laboratory and X-ray costs. For the secondary care sector inpatient and outpatient costs were based on diagnostic related group (DRG) purchase prices. No attempt was made to discount costs or benefits in the first instance, as this was would have been too complicated. This was a pragmatic approach aimed at gaining the understanding and acceptance of PBMA by clinical staff. Discounting would have been done at a later stage of the project, if it had developed further. The two RHA groups again jointly derived the method for choosing and ranking investment and disinvestment proposals, although a somewhat different means of agreeing the benefit criteria was adopted. In the SRHA exercise, the Advisory Group was initially asked to derive criteria themselves, whereas in the MRHA exercise a list of possible criteria was presented to and subjected to revision by the group. This latter approach saved time while not appearing to encroach to an objectionable degree on the autonomy of the Advisory Group.
2.2. Phase II This phase involved: 1. collection of detailed data on costs and benefits of options; 2. a second Advisory Group workshop to (a) consider evidence of costs and benefits, which was presented by Advisory Group members and invited speakers who were asked to become advocates for and against investments and disinvestments and (b) agree the final rankings of investments and disinvestments using derived prioritisation criteria and an anonymised voting system. Data were obtained upon which to base estimates of the additional cost of the specific investment proposals and existing cost data were obtained for disinvestment proposals. In addition to cost data, evidence on effectiveness and value for money, obtained from a literature search, was provided. This was complimented by clinical knowledge, which aligned literature findings to the local situation. After the final workshop the Advisory Groups were asked to comment on the overall success of the workshop and specifically on the PBMA process. In the Midland Health region an anonymised survey was conducted to gauge opinions and a similar evaluation was done in the Southern region.
3. Results Detailed data on costs and benefits are not reported here. Only background data on costs, criteria for judging benefits and final lists of priorities for investment and disinvestment are presented.
P. Bohmer et al. / Health Policy 55 (2001) 37–50
43
3.1. Background data on costs The estimated total respiratory disease budgets for the SRHA and MRHA were NZ$ 61 million and 58.5 million per annum, respectively.
3.2. Benefit criteria The lists of criteria agreed by the Advisory Groups are given in Table 3. These were used to aid judgements of the benefits provided by the different options. They were similar for both RHA groups, and are based primarily on the purchasing principles for regional health authorities contained in the Minister of Health’s Policy Guidelines [1]. The main difference is that the MRHA reserved some of the criteria for a further prioritisation process, which was applied to the proposals by the authority. Table 3 List of benefit criteria employed by Advisory Groups in reaching decisions about priority SRHA Advisory Group
MRHA Advisory Group
Evidence of effectiveness Efficiency Equity Risk management (non financial) Safety Acceptability
Evidence of effectiveness Efficiency Equity Health gain Disease burden (prevalence and severity) Social acceptabilitya Impact on priority groupsa
a
Criteria reserved by the MRHA.
3.3. In6estment and disin6estment proposals The final lists of investment and disinvestment proposals made by the Advisory Groups are given in Table 4. These two lists of investments and disinvestments are ranked in priority order. Proposed investments with the highest rankings have highest priority for investment. Proposed disinvestments with the highest ranking have highest priority for disinvestment.
3.4. Implementation This phase of the process developed somewhat differently in the following two regions.
3.4.1. Southern region Workshop participants did not want the SRHA to move ahead independently and implement any changes based on the list of options. Consequently, three
P. Bohmer et al. / Health Policy 55 (2001) 37–50
44
Table 4 Ranked investment and disinvestment proposals
Investments
Disinvestments
SRHA
MRHA
(i) Smoking prevention/cessation/lobbying and co-ordination (ii) Educate health professionals, the community and clients in respiratory disease issues (iii) Community based consultant clinics (iv) Sleep apnoea services (v) Introduction of the drug salmeterol
(i) Smoking cessation programmes
(i) Reduce inappropriate prescribing of cough medicine, antibiotics, inhaled steroids and bronchodilators (ii) Reduce follow-up OPD visits for people with respiratory conditions (iii) Reduce duplication of investigations (iv) No sputum investigations (both micro and cytology) (v) Lung transplants
(ii) Promote appropriate pharmaceutical use (iii) Extend flu immunisation programme (iv) Promotion of flu vaccine (v) Pneumococcal vaccine for high-risk groups (vi) Respiratory health educators (vii) Sleep apnoea services (viii) Home-based case management. Community based consultant clinics (i) Allergy testing
(ii) Lung transplants (iii) Radiotherapy and chemotherapy for lung cancer
separate projects were set up to investigate the three main areas for investment. The three projects were: 1. Stop smoking/prevent smoking uptake programmes; 2. Asthma/COPD education campaigns for GPs and consumers; and 3. Sleep disorder, and in particular sleep apnoea, services. It was intended that the groups would report back in time for recommendations to be built into the SRHA’s purchasing cycle for 1997/1998. This did not occur for a number of reasons, but the findings of the exercise influenced a number of the SRHA’s work areas and projects.
3.4.2. Midland region In the MRHA agreement was reached to explore the feasibility of implementing the various investment and disinvestment proposals but MRHA undertook to conduct further work on this. An internal working group was established to decide which were feasible and to develop implementation plans. The Executive approved implementation plans for: 1. Smoking cessation support through nicotine replacement therapies; 2. GP and patient education about antibiotic prescribing; and
P. Bohmer et al. / Health Policy 55 (2001) 37–50
45
3. Influenza vaccine promotion. These implementation plans were to be included in the 1997/1998 business plan for MRHA. New resources were to be found for implementation, meaning that no disinvestment needed to occur.
3.5. E6aluation The evaluation conducted by MRHA showed unanimous support among Advisory Group members for consulting with providers and consumers about prioritisation decisions and 90% of them felt that RHAs should be making judgements about the relative costs and benefits of health services. 4. Discussion This application of PBMA was considered a success by both RHAs, and implementation of the Advisory Groups’ recommendations is expected to lead to health gain for the people of both regions. However, there are a number of valuable lessons that have been learned.
4.1. Planning In both the RHAs there were specific reasons for choosing the respiratory diseases programme to apply PBMA. First, providers had been encouraging the RHAs to examine the possibility of establishing new sleep apnoea services and there was also support for introducing new expensive treatments for asthma, such as salmeterol. There were, therefore, a number of potential new treatments which providers wished to see evaluated. Secondly, the burden associated with respiratory diseases is substantial. They consume a significant proportion of health care resources and are an important cause of mortality and morbidity. This meant that there was likely to be interest in, and commitment to, taking part in an exercise of this kind. The implication is that the choice of programme area for applying PBMA was important to the success of the process. Assembling a suitable advisory group is another important part of the planning process for the application of PBMA. Advisory Group membership was based on possession of relevant knowledge and expertise, and balanced geographic, speciality, professional and lay group representation. Despite careful consideration of who should be invited, both the RHAs found that their original advisory groups were not comprehensive enough to provide relevant expertise. This may be inevitable, as it is the choice of specific investment and disinvestment proposals which makes particular expertise necessary during later stages of the decision making process. Therefore, it is probably reasonable to choose a broadly representative advisory group to make initial decisions about investment and disinvestment proposals and then to supplement this group with members possessing specific expertise in relevant areas for the final decision making stage. Deciding appropriate membership for the advisory groups is therefore critical to the success of the process.
46
P. Bohmer et al. / Health Policy 55 (2001) 37–50
4.2. Support Background work in preparation for the workshops was vital to the success of the process. The principal component of this preparation was the gathering of evidence to inform the Advisory Group’s decision making. For the first workshop, the burden of this work fell on RHA staff, but collaboration between the two RHA groups resulted in economies of effort. This collaboration was an important feature of the exercise and also provided the opportunity to compare the process and outcomes of the two RHA exercises. However, it needs to be recognised that a high level of involvement of staff with relevant skills is required throughout. For the second workshop, the knowledge and expertise of Advisory Group members was exploited to inform decisions, although RHA staff supported their information gathering efforts. Based on the evaluations of the process, one of the weaknesses perceived by participants was inadequacy of information. This lack of evidence to support decision making is a persistent problem in the health system and it is through explicit evidence based prioritisation frameworks, like PBMA, that such information gaps are identified. A further important ingredient in the success of the project was the development of a disease and intervention framework (Table 2). This grouped the range of existing or potential interventions into four broad categories, ranging from prevention through to continuing care and support, and facilitated the identification of the full range of possible interventions for each condition. The Advisory Groups worked through this framework and contributed their specific knowledge and experience to describe the range of appropriate interventions. This began the process of establishing ‘ownership’ of the exercise by the Advisory Groups.
4.3. Workshops An independent professional facilitator, with knowledge of the PBMA methodology and the health system, ran the two workshops. This independent facilitator was seen as impartial. The alternative approach of employing RHA staff, who were also involved in contract negotiations, as facilitators was felt likely to inhibit Advisory Group members’ participation in what became a remarkably non-partisan process. A key role of the facilitator is to ensure that there is clarity about the specific investment and disinvestment options and what they mean. Without this clarity final decision making and implementation is difficult. Clarity was achieved but a gradual process of refinement had to take place during the exercise. It is recommended that detailed descriptions of the investment and disinvestment options are recorded and agreed, subject to revision by the group, so that there is no remaining ambiguity. The approach of inviting Advisory Group members to become advocates for and against individual investment and disinvestment options gave them control over the process. This worked well and encouraged discussion and debate among participants. There was, however, an imbalance in the nature of advocacy for investment and disinvestment options. Advocates for investment and against disinvestment
P. Bohmer et al. / Health Policy 55 (2001) 37–50
47
were invited. This biased the process against disinvestment. This was felt to be a reasonable approach as it provided the opportunity for experts in the relevant fields to make a case against disinvestment, in what were highly contentious and controversial areas. Presentations on disinvestments lead to heated discussion. The anonymous voting system was important as it allowed Advisory Group members to vote freely. During the ranking exercise in the second and final workshop it was apparent that some members of the Advisory Group had found this very difficult. Also, discussion took place among the group about the merits of differentially weighting the criteria. In view of this difficulty and the discussion about weighting, the MRHA re-ranked the proposals using agreed weighted criteria. This led to some minor changes in the ranking order. In principle, weighted criteria are preferable because they are likely to more closely reflect the value judgements being placed on proposals than non-weighted (i.e. equally weighted) criteria. The simplest way of establishing weighting for criteria is to take the arithmetic mean of the weights assigned by group members.
4.4. Outcomes The list of investment and disinvestment proposals are interesting because the investment proposals are primarily preventive and non-clinical interventions. This might be considered surprising in view of the predominantly clinical composition of the groups. It is encouraging that clinicians did not promote narrow sectional interests but took a broader population-perspective on improving health gain. The other notable feature of the two lists of proposals is their similarity. For both groups, smoking cessation was the highest ranked proposal and three out of the five proposals suggested by the SRHA were also suggested by the MRHA group. Disinvestment proposals were less similar but there was also overlap here. One of the disinvestments suggested by the SRHA group (reduce inappropriate prescribing) is actually an investment which corresponds with one suggested by the MRHA group. The fact that two completely unconnected groups of health care professionals and patient advocates came up with very similar lists of proposals is encouraging and suggests that the process can produce consistent results. An important outcome was the demonstration of the value of co-operation between purchasers and providers and between purchasing authorities. The benefits of this exercise have been achieved by two RHAs working closely together, and with providers in a collaborative atmosphere. The evaluation demonstrated that this aspect of the process was highly valued by the participants.Informal feedback from both the SRHA and MRHA Advisory Groups showed overwhelming support for the process with members feeling it had been a worthwhile and valuable exercise. They were particularly supportive of the RHAs consulting providers about purchasing priorities. It is hoped that the good working relationships established during the conduct of this exercise will have enduring benefit. The benefits of PBMA must however be set against the costs. This is an expensive process. The MRHA exercise was estimated to have consumed a total of about 900 hours of working time by internal staff, Advisory Group members and other
48
P. Bohmer et al. / Health Policy 55 (2001) 37–50
contributors. This makes it a substantial investment. However, other alternative approaches to priority setting are also likely to be expensive and may not have the advantages of PBMA.
4.5. General lessons In this exercise a condition (disease), rather than a service based programme definition was chosen. This was an important decision. The advantage of this approach is that the continuum of effective interventions, across primary, secondary and tertiary care boundaries and between prevention and treatment, for each condition, could be examined. An alternative approach would be to define programmes in terms of services. This would tend to entrench existing boundaries between specialties and levels of care. Another method would be to define programmes in terms of eligible population groups, for example, children and the elderly. The framework adopted for this exercise (Figure 2) encouraged the Advisory Groups to take a broad integrated view of the range of possible interventions. This ‘service integration’ perspective is inherently encouraged by a condition-based framework. In conclusion, the perceived advantages of PBMA are that proposals for priorities are based on a rational, explicit, evidence-based process and carry with them the combined credibility and support of an expert advisory group. The process encourages co-operative working and may itself have enduring benefits. The disadvantages of this process, such as the costs and the requirement for skilled staff input, must be weighed against those of the alternatives. The comparative value of priorities derived by other means should also be considered. Implementation of the specific proposals has been relatively easy and this is believed to be partly due to the fact that they have been derived by an expert group, using the PBMA methodology, which has given them added credibility. There are also risks involved in conducting such an exercise. There is a potential for conflict among advisory group members and between the advisory group and the purchaser. This did not happen in this case and constructive relationships were maintained throughout. But the potential is there. Also, there is a risk that the purchasing authority will lose credibility if recommendations are not implemented. It is hoped that the planned extent of implementation of the recommendations for the 1997/98 year will meet with the approval of the Advisory Groups. PBMA has been strongly advocated as a method for prioritisation and resource allocation in the New Zealand health system [2] and other countries. The exercise described in this paper has been successful but has revealed some of the important practical issues that need to be addressed by those employing the PBMA methodology. It is hoped that the experience outline in this paper will enable those who use PBMA to do so effectively. It is interesting to note that one of the RHAs felt unable to disinvest in those areas proposed by the Advisory Group despite the recommendation to do so. This failure to pursue disinvestments does not represent a failure of the method or a
P. Bohmer et al. / Health Policy 55 (2001) 37–50
49
questioning of the validity of the Advisory Group’s recommendations but illustrates the difficulties encountered by purchasers in making such disinvestments. It is anticipated that purchasers in the New Zealand health system will be unable to avoid making such difficult decisions in the immediate future, as they will be compelled by their constrained resources to do so. The Health Funding Authority, which replaced the RHAs in 1998, is currently developing a framework and methods for prioritisation within the New Zealand Health System. Prioritisation methods, such as PBMA, which are explicitly evidence-based and rational, produce results which can be justified and which are most likely to achieve the aim of maximising health gain within available resources.
5. Summary: expert groups can ration rationally In New Zealand the rationing debate is moving on from theory into practice. Bohmer et al. have applied a health economics tool for priority setting and resource allocation to decision-making for the same service in two regional health authority areas. This has provided the opportunity to compare the decisions reached by two independent expert advisory groups comprised predominantly of clinicians but including management and community representatives. Both the groups gave high priority to investment in prevention and there was remarkable consistency between their specific investment and disinvestment proposals. This project has demonstrated that members of an expert panel can, for the most part, set aside sectional interests and agree evidence-based recommendations for maximising health gain within available resources.
Acknowledgements We would like to acknowledge the contributions of Advisory Group members who set aside sectional interests and gave their time and energy to this difficult task. We would also like to acknowledge the contributions of Jacqui Moore, Des O’Dea, Jit Cheung, Stephen Twitchin, Judy Katzenellenbogen and Jesse Kakou who assisted with the gathering of evidence on costs and benefits. Winston McKean provided valuable guidance and support to the project.
References [1] Shipley J. Policy Guidelines for Regional Health Authorities 1996/97, Ministry of Health, Wellington, 1996. [2] Brown MC, Cumming J. Development of a Conceptual Framework for RHA Purchasing: A Final Report to Midland Regional Health Authority, Health Services Research Centre and Wellington School of Medicine, Wellington, 1996. [3] Twaddle S, Walker A. Programme Budgeting and Marginal Analysis: Application Within Programmes to Assist Purchasing in Greater Glasgow Health Board, Health Policy, 1995;3391 – 105. .
50
P. Bohmer et al. / Health Policy 55 (2001) 37–50
[4] Vaithianathan R. Improving Health Purchasing: A Pilot in Public Health. North Health, Auckland, 1996. [5] Honigsbaum F, Richards J, Lockett R. Priority Setting in Action: Purchasing Dilemmas, Radcliffe Medical Press Ltd, Oxford, 1995. [6] Pain CH, Oatham P, Pierce M, et al. Integration: Need-based Service Definitions Discussion Paper, Health Manager, 1997;4(2)10–14.
.
Maximising health gain within available resources in the New Zealand public health system Paul Bohmer a,*, Charles Pain b, Alex Watt c, Paul Abernethy a, Janet Sceats d a
Southern Regional Health Authority, P.O. Box 3877, Christchurch, New Zealand b Midland Regional Health Authority, P.O. Box 1031, Hamilton, New Zealand c Southern Regional Health Authority, P.O. Box 5849, Dunedin, New Zealand d Midland Regional Health Authority, P.O. Box 1031, Hamilton, New Zealand Received 4 March 2000; accepted 3 July 2000
Abstract Objecti6e: to obtain intervention-condition-specific investment and disinvestment recommendations which optimise the potential for health gain from existing respiratory diseases resource expenditure; and to trial a health economics technique for this purpose, assessing its usefulness as a means of prioritising health services resource allocation. Design: a programme budgeting and marginal analysis (PBMA) exercise drawing upon the expertise of an advisory group of clinicians, managers and consumer advocates, supported by health authority staff. Setting: the Southern and Midland health regions in New Zealand which have populations of one million and seven hundred thousand respectively. Health system context: publicly funded secondary care sector in which regional health authority (RHA) purchasers contract for services with health care providers. Methods: available evidence on the marginal costs and benefits of services for respiratory diseases was examined by an advisory group who produced investment and disinvestment recommendations by consensus using agreed prioritisation criteria. Results: A list of specific investment and disinvestment proposals. Implementation plans for a number of investments formed part of the business plans for both RHAs in 1997/1998. No disinvestments were planned. Conclusions: prioritisation methods like -PBMA, which are explicit and rational, can produce defensible evidence-based recommenda-
* Corresponding author. Present address: Health Funding Authority, Private Bag 92-522, Wellesley Street, Auckland, New Zealand. Tel.: +64-9-5809057; fax: +64-9-5809001. E-mail address: [email protected] (P. Bohmer). 0168-8510/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 5 1 0 ( 0 0 ) 0 0 1 0 7 - X
38
P. Bohmer et al. / Health Policy 55 (2001) 37–50
tions with the additional benefit of the credibility and support of an expert advisory group. The process encourages co-operative working and may itself have enduring benefits. However, preparation and conduct of such exercises is resource intensive and requires careful planning. This exercise has provided valuable lessons for the conduct of future prioritisation work. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: PBMA; Prioritisation; Secondary care; Respiratory services; Purchasers; New Zealand
1. Introduction This paper describes a prioritisation exercise conducted by two purchasers in the New Zealand Health System. This has provided a unique opportunity to compare the outcome of two such exercises in two different localities but in the same health service programme area. The methods and results are briefly described but the main emphasis is on discussion of the practical lessons learned, which may be valuable in future applications of this method.
1.1. Health system context In New Zealand, as in other health systems, growing demand for services has placed increasing pressure on health care resources. This has led to an emphasis on obtaining ‘value for money’ and an awareness of the opportunity costs of spending scarce resources on marginally effective or even harmful interventions. The regional health authorities (RHAs) were established with the implementation of the New Zealand Health Reforms in 1993 and were funded by means of a population-based funding formula (PBFF) to purchase public health (health promotion and prevention), personal health (family health and hospital services) and disability support services for the residents of their respective regions. In 1996, most New Zealand health services were purchased on a regional basis by one of four RHAs. These RHAs were responsible for improving, maintaining and protecting health [1]. Their role was to take an overview of the population’s need for health services and to optimise the effectiveness, efficiency, equity, acceptability, safety and risk management of service provision. A key part of this role was to prioritise services and interventions in order to maximise health gain for the populations served. However, much of the purchasing function has been carried out on a ‘role-over’ basis, with minimal change to the pattern of service provision. A means of making rational prioritisation decisions within an explicit cost-benefit framework is needed to enable the purchasers to make decisions which optimise health gain within the existing resource constraints. An economics-based model such as programme budgeting and marginal analysis (PBMA) offers a possible method for achieving this. And because PBMA takes into account needs, costs and benefits, it offers an advantage over needs assessment alone by considering outcomes and costs of treatment services in making prioritisation decisions.
P. Bohmer et al. / Health Policy 55 (2001) 37–50
39
1.2. PBMA PBMA is a method, which has been used in New Zealand and in other health systems, which involves evaluation of the costs and benefits of existing and potential services or interventions [2–5]. It comprises two activities: 1. programme budgeting, examining evidence on how the current resources of a defined programme are allocated; and 2. marginal analysis, which involves an evidence-based evaluation of the service mix to determine whether greater benefits could be realised by changing the proportionate allocations of resources among the various services or interventions. This involves making value judgements, supported by evidence, of whether greater benefits for the population could be achieved with the same resources by disinvesting in less effective services and reinvesting in more effective services. A framework for decision-making is thus provided in which value judgements are explicit and informed by evidence. An advisory group, comprised of individuals with relevant expertise and experience in the programme area and others who represent various interest groups, can be employed to make prioritisation decisions using this framework while applying their own value-judgements. For more details on PBMA, the reader is referred to Health Policy, 35 (2) (1995), which is entirely dedicated to PBMA.
1.3. Project background The Midland and Southern regional health authorities (MRHA and SRHA) decided to implement a project to assess the feasibility of applying PBMA as a means of prioritising resource allocation within an existing budget, and to obtain specific recommendations regarding the optimal allocation of resources in the important programme area of respiratory diseases. Respiratory diseases consume a significant proportion of health care resources. For example, in the Midland Health region, respiratory diseases consume approximately 7% of the total expenditure. The burden of these diseases is considerable. They account for some 7.2% of hospital admissions per year in the Southern region and 9% of admissions in the Midland region. They also account for some 13% of all the deaths in the Southern region and 16% of deaths in the Midland region. From the RHA perspective, therefore, they provided an opportunity to employ PBMA in an important area with the possibility of increasing health gain for patients within existing resources. 2. Methods The procedure for applying the PBMA methodology to the respiratory programme was developed jointly by representatives from the SRHA and MRHA, and involved an advisory group taking part in two workshops. Fig. 1 illustrates the PBMA procedure employed.
40
P. Bohmer et al. / Health Policy 55 (2001) 37–50
Fig. 1. PBMA process.
2.1. Phase I International classification of disease (ICD-9) codes were used to define the diseases included in the programme. This approach defined the respiratory programme in a way which ‘cut across’ the currently purchased respiratory and other programmes, for both of the RHAs. For example, lung cancer is currently included in the ‘oncology programme’ and not the ‘respiratory programme’ of both the Authorities. This broad condition-based approach to programme definition was adopted by both RHAs for consistency and because it is in keeping with the logic of an integrated disease management approach [6]. The conditions included in the respiratory programme for this exercise are given in Table 1. An inclusive programme budget was derived for all these conditions, comprising health promotion and prevention, primary health care, outpatients and secondary care, and laboratory and pharmaceutical costs. There was some difficulty in establishing these budgets as much of the available data was inadequate, particularly in the primary care sector. However, costs were based on best estimates and reflected the purchase costs of the services to the health authorities. Though they did not necessarily reflect actual hospital costs, they were likely to have been reasonable estimates. Other important activities in Phase I included:
P. Bohmer et al. / Health Policy 55 (2001) 37–50
41
Table 1 The respiratory programme as defined by ICD-9 codes Diseases
ICD 9 Classification
Tuberculosis Lung and pleural cancer Cystic fibrosis Acute bronchitis and bronchiolitis Pneumonia and Influenza COPD and allied conditions Asthma Pneumoconiosis Other diseases of the respiratory system Sleep apnoea and other sleep disturbances
11 162 277 466 480–486 491, 492, 494, 496 493 501.506, 507 510–518 780.5
1. identifying and inviting an Advisory Group with members chosen to provide representation from across each region and relevant expertise in the various health care disciplines involved in respiratory diseases management and care including respiratory physicians, ENT surgeons, general practitioners, oncologists, nursing staff, physiotherapists, hospital managers, academics, consumer groups, lay community members and representatives for Maori (the Advisory Groups in both the RHAs comprised of approximately one-third respiratory physicians and the majority of members in both groups were doctors); 2. collection of background data to provide an overview of disease burden and costs related to the respiratory programme; and 3. an Advisory Group workshop to produce a detailed description of the existing respiratory disease programme, discuss and identify potential candidates for investment or disinvestment and agree benefit criteria to be used in the marginal analysis. At the first workshop a disease and intervention framework was used to identify and list the full range of existing and potential interventions for each stage of all diseases in the respiratory programme. This framework was developed for the exercise and is shown in Table 2. Interventions were grouped under the headings of prevention, assessment and investigation, treatment, continuing care, support and rehabilitation. Table 2 Intervention/disease framework Type of intervention Tuberculosis Lung Cystic Pneumonia Acute COPD Etcetra cancer fibrosis and bronchitis and influenza bronchiolitis Prevention Assessment and investigation Treatment Continuing care, support, rehabilitation
42
P. Bohmer et al. / Health Policy 55 (2001) 37–50
Cost data were obtained for the existing respiratory programmes in each RHA. These data were collected from a number of sources. For public health an estimate was made of the costs of health education, smoking cessation programmes and health education for asthma. In primary care these costs consisted of both RHA subsidies and private fees paid for a GP consultation, pharmaceutical subsidy costs and laboratory and X-ray costs. For the secondary care sector inpatient and outpatient costs were based on diagnostic related group (DRG) purchase prices. No attempt was made to discount costs or benefits in the first instance, as this was would have been too complicated. This was a pragmatic approach aimed at gaining the understanding and acceptance of PBMA by clinical staff. Discounting would have been done at a later stage of the project, if it had developed further. The two RHA groups again jointly derived the method for choosing and ranking investment and disinvestment proposals, although a somewhat different means of agreeing the benefit criteria was adopted. In the SRHA exercise, the Advisory Group was initially asked to derive criteria themselves, whereas in the MRHA exercise a list of possible criteria was presented to and subjected to revision by the group. This latter approach saved time while not appearing to encroach to an objectionable degree on the autonomy of the Advisory Group.
2.2. Phase II This phase involved: 1. collection of detailed data on costs and benefits of options; 2. a second Advisory Group workshop to (a) consider evidence of costs and benefits, which was presented by Advisory Group members and invited speakers who were asked to become advocates for and against investments and disinvestments and (b) agree the final rankings of investments and disinvestments using derived prioritisation criteria and an anonymised voting system. Data were obtained upon which to base estimates of the additional cost of the specific investment proposals and existing cost data were obtained for disinvestment proposals. In addition to cost data, evidence on effectiveness and value for money, obtained from a literature search, was provided. This was complimented by clinical knowledge, which aligned literature findings to the local situation. After the final workshop the Advisory Groups were asked to comment on the overall success of the workshop and specifically on the PBMA process. In the Midland Health region an anonymised survey was conducted to gauge opinions and a similar evaluation was done in the Southern region.
3. Results Detailed data on costs and benefits are not reported here. Only background data on costs, criteria for judging benefits and final lists of priorities for investment and disinvestment are presented.
P. Bohmer et al. / Health Policy 55 (2001) 37–50
43
3.1. Background data on costs The estimated total respiratory disease budgets for the SRHA and MRHA were NZ$ 61 million and 58.5 million per annum, respectively.
3.2. Benefit criteria The lists of criteria agreed by the Advisory Groups are given in Table 3. These were used to aid judgements of the benefits provided by the different options. They were similar for both RHA groups, and are based primarily on the purchasing principles for regional health authorities contained in the Minister of Health’s Policy Guidelines [1]. The main difference is that the MRHA reserved some of the criteria for a further prioritisation process, which was applied to the proposals by the authority. Table 3 List of benefit criteria employed by Advisory Groups in reaching decisions about priority SRHA Advisory Group
MRHA Advisory Group
Evidence of effectiveness Efficiency Equity Risk management (non financial) Safety Acceptability
Evidence of effectiveness Efficiency Equity Health gain Disease burden (prevalence and severity) Social acceptabilitya Impact on priority groupsa
a
Criteria reserved by the MRHA.
3.3. In6estment and disin6estment proposals The final lists of investment and disinvestment proposals made by the Advisory Groups are given in Table 4. These two lists of investments and disinvestments are ranked in priority order. Proposed investments with the highest rankings have highest priority for investment. Proposed disinvestments with the highest ranking have highest priority for disinvestment.
3.4. Implementation This phase of the process developed somewhat differently in the following two regions.
3.4.1. Southern region Workshop participants did not want the SRHA to move ahead independently and implement any changes based on the list of options. Consequently, three
P. Bohmer et al. / Health Policy 55 (2001) 37–50
44
Table 4 Ranked investment and disinvestment proposals
Investments
Disinvestments
SRHA
MRHA
(i) Smoking prevention/cessation/lobbying and co-ordination (ii) Educate health professionals, the community and clients in respiratory disease issues (iii) Community based consultant clinics (iv) Sleep apnoea services (v) Introduction of the drug salmeterol
(i) Smoking cessation programmes
(i) Reduce inappropriate prescribing of cough medicine, antibiotics, inhaled steroids and bronchodilators (ii) Reduce follow-up OPD visits for people with respiratory conditions (iii) Reduce duplication of investigations (iv) No sputum investigations (both micro and cytology) (v) Lung transplants
(ii) Promote appropriate pharmaceutical use (iii) Extend flu immunisation programme (iv) Promotion of flu vaccine (v) Pneumococcal vaccine for high-risk groups (vi) Respiratory health educators (vii) Sleep apnoea services (viii) Home-based case management. Community based consultant clinics (i) Allergy testing
(ii) Lung transplants (iii) Radiotherapy and chemotherapy for lung cancer
separate projects were set up to investigate the three main areas for investment. The three projects were: 1. Stop smoking/prevent smoking uptake programmes; 2. Asthma/COPD education campaigns for GPs and consumers; and 3. Sleep disorder, and in particular sleep apnoea, services. It was intended that the groups would report back in time for recommendations to be built into the SRHA’s purchasing cycle for 1997/1998. This did not occur for a number of reasons, but the findings of the exercise influenced a number of the SRHA’s work areas and projects.
3.4.2. Midland region In the MRHA agreement was reached to explore the feasibility of implementing the various investment and disinvestment proposals but MRHA undertook to conduct further work on this. An internal working group was established to decide which were feasible and to develop implementation plans. The Executive approved implementation plans for: 1. Smoking cessation support through nicotine replacement therapies; 2. GP and patient education about antibiotic prescribing; and
P. Bohmer et al. / Health Policy 55 (2001) 37–50
45
3. Influenza vaccine promotion. These implementation plans were to be included in the 1997/1998 business plan for MRHA. New resources were to be found for implementation, meaning that no disinvestment needed to occur.
3.5. E6aluation The evaluation conducted by MRHA showed unanimous support among Advisory Group members for consulting with providers and consumers about prioritisation decisions and 90% of them felt that RHAs should be making judgements about the relative costs and benefits of health services. 4. Discussion This application of PBMA was considered a success by both RHAs, and implementation of the Advisory Groups’ recommendations is expected to lead to health gain for the people of both regions. However, there are a number of valuable lessons that have been learned.
4.1. Planning In both the RHAs there were specific reasons for choosing the respiratory diseases programme to apply PBMA. First, providers had been encouraging the RHAs to examine the possibility of establishing new sleep apnoea services and there was also support for introducing new expensive treatments for asthma, such as salmeterol. There were, therefore, a number of potential new treatments which providers wished to see evaluated. Secondly, the burden associated with respiratory diseases is substantial. They consume a significant proportion of health care resources and are an important cause of mortality and morbidity. This meant that there was likely to be interest in, and commitment to, taking part in an exercise of this kind. The implication is that the choice of programme area for applying PBMA was important to the success of the process. Assembling a suitable advisory group is another important part of the planning process for the application of PBMA. Advisory Group membership was based on possession of relevant knowledge and expertise, and balanced geographic, speciality, professional and lay group representation. Despite careful consideration of who should be invited, both the RHAs found that their original advisory groups were not comprehensive enough to provide relevant expertise. This may be inevitable, as it is the choice of specific investment and disinvestment proposals which makes particular expertise necessary during later stages of the decision making process. Therefore, it is probably reasonable to choose a broadly representative advisory group to make initial decisions about investment and disinvestment proposals and then to supplement this group with members possessing specific expertise in relevant areas for the final decision making stage. Deciding appropriate membership for the advisory groups is therefore critical to the success of the process.
46
P. Bohmer et al. / Health Policy 55 (2001) 37–50
4.2. Support Background work in preparation for the workshops was vital to the success of the process. The principal component of this preparation was the gathering of evidence to inform the Advisory Group’s decision making. For the first workshop, the burden of this work fell on RHA staff, but collaboration between the two RHA groups resulted in economies of effort. This collaboration was an important feature of the exercise and also provided the opportunity to compare the process and outcomes of the two RHA exercises. However, it needs to be recognised that a high level of involvement of staff with relevant skills is required throughout. For the second workshop, the knowledge and expertise of Advisory Group members was exploited to inform decisions, although RHA staff supported their information gathering efforts. Based on the evaluations of the process, one of the weaknesses perceived by participants was inadequacy of information. This lack of evidence to support decision making is a persistent problem in the health system and it is through explicit evidence based prioritisation frameworks, like PBMA, that such information gaps are identified. A further important ingredient in the success of the project was the development of a disease and intervention framework (Table 2). This grouped the range of existing or potential interventions into four broad categories, ranging from prevention through to continuing care and support, and facilitated the identification of the full range of possible interventions for each condition. The Advisory Groups worked through this framework and contributed their specific knowledge and experience to describe the range of appropriate interventions. This began the process of establishing ‘ownership’ of the exercise by the Advisory Groups.
4.3. Workshops An independent professional facilitator, with knowledge of the PBMA methodology and the health system, ran the two workshops. This independent facilitator was seen as impartial. The alternative approach of employing RHA staff, who were also involved in contract negotiations, as facilitators was felt likely to inhibit Advisory Group members’ participation in what became a remarkably non-partisan process. A key role of the facilitator is to ensure that there is clarity about the specific investment and disinvestment options and what they mean. Without this clarity final decision making and implementation is difficult. Clarity was achieved but a gradual process of refinement had to take place during the exercise. It is recommended that detailed descriptions of the investment and disinvestment options are recorded and agreed, subject to revision by the group, so that there is no remaining ambiguity. The approach of inviting Advisory Group members to become advocates for and against individual investment and disinvestment options gave them control over the process. This worked well and encouraged discussion and debate among participants. There was, however, an imbalance in the nature of advocacy for investment and disinvestment options. Advocates for investment and against disinvestment
P. Bohmer et al. / Health Policy 55 (2001) 37–50
47
were invited. This biased the process against disinvestment. This was felt to be a reasonable approach as it provided the opportunity for experts in the relevant fields to make a case against disinvestment, in what were highly contentious and controversial areas. Presentations on disinvestments lead to heated discussion. The anonymous voting system was important as it allowed Advisory Group members to vote freely. During the ranking exercise in the second and final workshop it was apparent that some members of the Advisory Group had found this very difficult. Also, discussion took place among the group about the merits of differentially weighting the criteria. In view of this difficulty and the discussion about weighting, the MRHA re-ranked the proposals using agreed weighted criteria. This led to some minor changes in the ranking order. In principle, weighted criteria are preferable because they are likely to more closely reflect the value judgements being placed on proposals than non-weighted (i.e. equally weighted) criteria. The simplest way of establishing weighting for criteria is to take the arithmetic mean of the weights assigned by group members.
4.4. Outcomes The list of investment and disinvestment proposals are interesting because the investment proposals are primarily preventive and non-clinical interventions. This might be considered surprising in view of the predominantly clinical composition of the groups. It is encouraging that clinicians did not promote narrow sectional interests but took a broader population-perspective on improving health gain. The other notable feature of the two lists of proposals is their similarity. For both groups, smoking cessation was the highest ranked proposal and three out of the five proposals suggested by the SRHA were also suggested by the MRHA group. Disinvestment proposals were less similar but there was also overlap here. One of the disinvestments suggested by the SRHA group (reduce inappropriate prescribing) is actually an investment which corresponds with one suggested by the MRHA group. The fact that two completely unconnected groups of health care professionals and patient advocates came up with very similar lists of proposals is encouraging and suggests that the process can produce consistent results. An important outcome was the demonstration of the value of co-operation between purchasers and providers and between purchasing authorities. The benefits of this exercise have been achieved by two RHAs working closely together, and with providers in a collaborative atmosphere. The evaluation demonstrated that this aspect of the process was highly valued by the participants.Informal feedback from both the SRHA and MRHA Advisory Groups showed overwhelming support for the process with members feeling it had been a worthwhile and valuable exercise. They were particularly supportive of the RHAs consulting providers about purchasing priorities. It is hoped that the good working relationships established during the conduct of this exercise will have enduring benefit. The benefits of PBMA must however be set against the costs. This is an expensive process. The MRHA exercise was estimated to have consumed a total of about 900 hours of working time by internal staff, Advisory Group members and other
48
P. Bohmer et al. / Health Policy 55 (2001) 37–50
contributors. This makes it a substantial investment. However, other alternative approaches to priority setting are also likely to be expensive and may not have the advantages of PBMA.
4.5. General lessons In this exercise a condition (disease), rather than a service based programme definition was chosen. This was an important decision. The advantage of this approach is that the continuum of effective interventions, across primary, secondary and tertiary care boundaries and between prevention and treatment, for each condition, could be examined. An alternative approach would be to define programmes in terms of services. This would tend to entrench existing boundaries between specialties and levels of care. Another method would be to define programmes in terms of eligible population groups, for example, children and the elderly. The framework adopted for this exercise (Figure 2) encouraged the Advisory Groups to take a broad integrated view of the range of possible interventions. This ‘service integration’ perspective is inherently encouraged by a condition-based framework. In conclusion, the perceived advantages of PBMA are that proposals for priorities are based on a rational, explicit, evidence-based process and carry with them the combined credibility and support of an expert advisory group. The process encourages co-operative working and may itself have enduring benefits. The disadvantages of this process, such as the costs and the requirement for skilled staff input, must be weighed against those of the alternatives. The comparative value of priorities derived by other means should also be considered. Implementation of the specific proposals has been relatively easy and this is believed to be partly due to the fact that they have been derived by an expert group, using the PBMA methodology, which has given them added credibility. There are also risks involved in conducting such an exercise. There is a potential for conflict among advisory group members and between the advisory group and the purchaser. This did not happen in this case and constructive relationships were maintained throughout. But the potential is there. Also, there is a risk that the purchasing authority will lose credibility if recommendations are not implemented. It is hoped that the planned extent of implementation of the recommendations for the 1997/98 year will meet with the approval of the Advisory Groups. PBMA has been strongly advocated as a method for prioritisation and resource allocation in the New Zealand health system [2] and other countries. The exercise described in this paper has been successful but has revealed some of the important practical issues that need to be addressed by those employing the PBMA methodology. It is hoped that the experience outline in this paper will enable those who use PBMA to do so effectively. It is interesting to note that one of the RHAs felt unable to disinvest in those areas proposed by the Advisory Group despite the recommendation to do so. This failure to pursue disinvestments does not represent a failure of the method or a
P. Bohmer et al. / Health Policy 55 (2001) 37–50
49
questioning of the validity of the Advisory Group’s recommendations but illustrates the difficulties encountered by purchasers in making such disinvestments. It is anticipated that purchasers in the New Zealand health system will be unable to avoid making such difficult decisions in the immediate future, as they will be compelled by their constrained resources to do so. The Health Funding Authority, which replaced the RHAs in 1998, is currently developing a framework and methods for prioritisation within the New Zealand Health System. Prioritisation methods, such as PBMA, which are explicitly evidence-based and rational, produce results which can be justified and which are most likely to achieve the aim of maximising health gain within available resources.
5. Summary: expert groups can ration rationally In New Zealand the rationing debate is moving on from theory into practice. Bohmer et al. have applied a health economics tool for priority setting and resource allocation to decision-making for the same service in two regional health authority areas. This has provided the opportunity to compare the decisions reached by two independent expert advisory groups comprised predominantly of clinicians but including management and community representatives. Both the groups gave high priority to investment in prevention and there was remarkable consistency between their specific investment and disinvestment proposals. This project has demonstrated that members of an expert panel can, for the most part, set aside sectional interests and agree evidence-based recommendations for maximising health gain within available resources.
Acknowledgements We would like to acknowledge the contributions of Advisory Group members who set aside sectional interests and gave their time and energy to this difficult task. We would also like to acknowledge the contributions of Jacqui Moore, Des O’Dea, Jit Cheung, Stephen Twitchin, Judy Katzenellenbogen and Jesse Kakou who assisted with the gathering of evidence on costs and benefits. Winston McKean provided valuable guidance and support to the project.
References [1] Shipley J. Policy Guidelines for Regional Health Authorities 1996/97, Ministry of Health, Wellington, 1996. [2] Brown MC, Cumming J. Development of a Conceptual Framework for RHA Purchasing: A Final Report to Midland Regional Health Authority, Health Services Research Centre and Wellington School of Medicine, Wellington, 1996. [3] Twaddle S, Walker A. Programme Budgeting and Marginal Analysis: Application Within Programmes to Assist Purchasing in Greater Glasgow Health Board, Health Policy, 1995;3391 – 105. .
50
P. Bohmer et al. / Health Policy 55 (2001) 37–50
[4] Vaithianathan R. Improving Health Purchasing: A Pilot in Public Health. North Health, Auckland, 1996. [5] Honigsbaum F, Richards J, Lockett R. Priority Setting in Action: Purchasing Dilemmas, Radcliffe Medical Press Ltd, Oxford, 1995. [6] Pain CH, Oatham P, Pierce M, et al. Integration: Need-based Service Definitions Discussion Paper, Health Manager, 1997;4(2)10–14.
.