Skip to main content
European Commission logo
Deutsch Deutsch
CORDIS - Forschungsergebnisse der EU
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary
Inhalt archiviert am 2024-06-18

Operations management and demand-based approaches to healthcare outcomes and cost-benefits research

Final Report Summary - MANAGED OUTCOMES (Operations management and demand-based approaches to healthcare outcomes and cost-benefits research)

Executive summary:

The project MANAGED OUTCOMES (Operations Management and Demand-based approaches to Healthcare Outcomes and Cost-benefits Research) is co-funded by the European Seventh Framework Programme in the HEALTH priority. The project has addressed the need for better understanding the relationships between health outcomes and cost-benefits and how they are affected by service production efficiency and by the regional structure of healthcare supply chains. The studies performed in the project will support more optimized health systems and contribute to better healthcare planning.

The project, which began in January 2010 and ran to December 2012, has been carried out by a consortium comprising ten organizations from eight different European countries. The project focussed to start with building a strong methodological basis for the work and then moved on to collecting data from 24 case instances around Europe. These data were collected from six countries with different types of healthcare systems: Finland, Germany, Greece, the Netherlands, Spain and the UK. Four large patient groups, each representing different health service demand types, were selected: type 2 diabetes, stroke, hip osteoarthritis and dementia.

Type 2 diabetes is a high volume and high cost disease where early stage detection and treatment efforts pay off in terms of providing more cost-effective care. In the MANAGED OUTCOMES project easy access to, and coordination by, primary health care came out as important factors positively affecting health status as reported by patients. In addition, as there are different stages in the disease and service requirements vary, demand–based segmentation helps in understanding and studying the provision of care. The future scenarios for type 2 diabetes included low cost possibilities as well as scenarios where diabetic control is maintained and complications are minimized.

There were large variations identified in the organisation of stroke care services in different regions. Progress towards widely accepted care guidelines is still uneven across Europe and this seems to contribute to variation in health outcomes. A focus on faster access to specialist stroke services and earlier identification of symptoms by patients and the public are seen as important scenarios for the future of stroke services development.

Hip osteoarthritis – and its curative treatment through surgery - represented an elective service in the project. Key findings included evidence of the positive effects of centralised and high volume treatment facilities for cost-efficiency. There were wide variations in the incidence of hip replacements and total regional costs seem to be driven by this rather than process efficiency. The study indicates that the incidence of operations is partly related to the availability of resources but that systematic demand management could play a bigger role in the future organization of care.

As dementia diagnoses tend to be poorly recorded in information systems in all regions, and many different service providers are involved, it was the most complex case study we chose to examine. Despite this the MANAGED OUTCOMES project identified large differences in the organisation of care in various European regions and effects of community involvement and early diagnosis. The future scenarios identified different care configurations that could provide the coordination needed to progress the integration of dementia care.

The MANAGED OUTCOMES approach was successful in developing a rich picture of the functioning and performance of regional service delivery systems for specific patient groups. The comparative analysis allowed us to compare different practices across Europe and identify the potential for learning and improving services, despite the differences in healthcare systems and cultures. The multi-perspective approach resulted in many interesting conclusions to improve healthcare practice by developing more effective and efficient health service delivery systems.

The recommendations of MANAGED OUTCOMES emphasise the need for focus in planning and controlling regional health service provision systems and offer a methodology to support this. We believe that better understanding of the demand for services– and consistent development of demand management practices in healthcare - may help overcome the future health challenges of providing appropriate services for ageing populations and ease pressures on healthcare funding.

Detailed information on the project results, the team and all public reports are available on the MANAGED OUTCOMES website http://www.managedoutcomes.eu.

Project Context and Objectives:

CONTEXT
MANAGED OUTCOMES is a collaborative project co-funded by the 7th Framework Programme. It explores the assumption that healthcare outcomes and costs are affected by the efficiency of service production, the time/location constrained regional structure of healthcare delivery, and the degree to which people are able to participate in the co-production of their care. These relationships are insufficiently understood and need to be studied in order to meet the objectives of the new European health strategy. Better healthcare systems can then be planned to meet citizens' expectations for improved access to, and quality of, healthcare, and hence improved outcomes.

As European healthcare systems are diverse, models have been created drawing from comparative analyses of treatment practices across Europe. Together with studies of economic, allocative and technical efficiency, these models have aimed to identify and illustrate good practices across case studies in selected European healthcare systems. Scenario methods were then applied to these models to outline potential future development paths for healthcare systems over the next decade. The studies have supported a better understanding of relationships between the costs and outcomes of healthcare systems and the development of tools to support policy and decision makers in healthcare system reforms.

THEORETICAL BACKGROUND
The project was based on three conceptual constructs drawn from the Operations Management and Service Science bodies of knowledge. These are 1) the separation of outputs and outcomes; 2) service delivery and distribution in economic regions; and 3) the necessity to segment demand and organize supply to meet it.

1) SEPARATION OF OUTPUTS AND OUTCOMES
Outputs are identifiable and measurable activities done to a patient, such as giving advice, a visit, prescription or surgical intervention. Outcomes are observable changes in a patient's health condition, i.e. what happens to a patient. Health outcomes can be understood as a sum of different contributions, of which the outputs produced by the healthcare system obviously are crucial. Other contributions may be placebo effects, patients' health behaviour, trust, and random events. In addition to the clinical quality of outputs, their efficacy is affected by their availability (time and location), as well as at what point in an episode of illness service supply meets demand. To improve healthcare production systems in the terms of its effects on health outcomes, the relationship between output contributions and realized outcomes has to be isolated. The crucial point, much emphasized in the Service Science literature, is that outcomes, and the ensuing health value, is jointly co-created by providers and patients. The purpose of a health service system is not to maximize the volume of outputs, but to maximize the accomplishment of outcomes, given the available resources.

2) SERVICE DELIVERY AND DISTRIBUTION IN ECONOMIC REGIONS
Assuming that the ways health services are made available to population impacts outcomes, the structure of service delivery and distribution needs to be studied. The key concepts here are demand-supply chains, which may be configured as supply stars (where all service delivery points are connected to one central supply hub), or networks (several interlinked nodes). Distribution systems are usually analyzed as contact points and stock keeping units (SKU), their catchment areas, and supply structures. In healthcare the equivalent is a Service Provision Point (SPP), i.e. a location or a contact point to where patients go to get help. The patient's visit or visits to different SPPs constitutes a patient journey, to which service providers need to organize care pathways, i.e. ways and means to enable a smooth and effective patient journey.

A SPP has three crucial attributes, access, variety, and specialization. Access, variety, and specialization constitute a trilemma, a three-factor optimization problem to which there is no optimal general solution. Any solution is constrained by capacity utilization and unit cost.

To the extent that patient journeys require different types and specialization levels of services, the care pathways need to be organized regionally. The level of the highest required specialization determines (theoretically), the catchment area of the care pathway. Therefore health service delivery and distribution systems need to be studied on a regional level. In practice a region is an economic unit, with a certain density of transactions and commuting. From a healthcare production perspective, a region consists of a set of service provision points (SPP) which can deliver care to most health problems, excluding very rare, complex, or costly cases, that call for supra-regional services (i.e. university hospitals).

In sum, healthcare service producers have to, under a combination of constraints, take decisions about the structure, location, and combinations of SPP's in order to meet patient demand. From a theoretical perspective it can be assumed that there is no (easy) optimal solution to the trilemma of access, variety, and specialization under the constraint of unit cost. From an empirical perspective it can be concluded that no universal one-best-way –solution has emerged, or made public. Therefore the research setting of MANAGED OUTCOMES was to, first, study how care pathways are organized in European regions and, second, study how the various solutions impact health outcomes.

3) DEMAND SEGMENTATION
Health service delivery and distribution systems appear to be very complex. However, complexity can be reduced or absorbed in many ways. A common way, widely applied in marketing, is segmentation. A population, a market, or a demand can be analytically and empirically divided into homogeneous subgroups following some classification. The segments are, in some important ways, sufficiently homogeneous. Thereby the supply side can develop specialized solutions to each segment.

A general assumption is that healthcare is an industry with a single, but complex, industry logic. In MANAGED OUTCOMES we have rejected this assumption. Instead the proposition is that healthcare is a cluster of industries, each with its own operating and economic logic. The participating industries cooperate around a common core, helping people with medical problems, while each uses different technologies, business models and operating principles. Such logics can be used as a basis for segmentation.

While the demographic, urgency, and clinical categories are relevant and cannot be overlooked, MANAGED OUTCOMES applies a demand-supply-based operating logic (DSO) as a segmentation principle. The DSO model is based on the assumption that demand and supply conditions together require that service production takes on a specific operating logic. In general, health services cannot produce everything demanded; it is constrained by what is technically possible, patients' health behaviour and compliance, and the availability of resources. Demand for health services can be classified with the following variables:
- urgent vs. not urgent
- severe vs. not severe
- health-related risks vs. actual disease or wound.

A crucial issue is how supply meets demand. An Order Penetration Point (OPP) signifies the point in time and space where a specific customer request meets or activates a specific supply. In healthcare the OPP can be seen as a point on the time-line of an episode of illness, i.e. the point at which a patient journey starts (a person becomes a patient). The OPP can be activated in different ways at different times, depending on the demand-supply situation. For example, screening and prevention has the aim of moving the OPP to an earlier point in time.

Combining the supply constraints with basic demand types gives seven DSO-logics:
1) Prevention deals with situations where patients have an elevated risk, but do not yet suffer from significant discomfort or pain. What is typically required is a change in life-style, meaning that the principal actor is the patient, with some assistance and support of the service provider.
2) Emergency deals with urgent and severe issues, where there is a time-window to save and stabilize. Patients typically can self-select and are motivated to treatment. Emergency services meet unsorted demand as it arises; therefore it requires stand-by capacity, and a large variety of capabilities.
3) One-visit deals with non-urgent, non-severe issues, which, if the production system allows, can be treated during one visit and one continuous workflow. Previous medical history is not highly relevant and the visit closes the case.
4) Electives are processes where a precision diagnosis results in a precision procedure that can be planned and scheduled. The intervention typically leads to a stepwise change in the patient's condition.
5) Cure processes deal with situations where a reliable and precise diagnosis and a schedulable care plan can't be developed, as the case involves uncertainties. Therefore the process is emergent, i.e. it can be planned only a few steps ahead, after which the patient's response to treatment needs to be assessed and plans readjusted. The assumption, though, is that the patient's ailment can be cured or eased.
6) Continuous care deals with chronic or terminal conditions to which there is no known cure and thereby no desirable end point. Therefore the aim is to maintain quality of life and arrest decline. Care processes require a regular rhythm of monitoring and therapy.
7) Projects deal with cases that are complex and rare to the extent that no pre-planned processes can be applied, but each patient needs to be treated as a unique project, preferably with a care manager that orchestrates various procedures, examinations and evaluations.

The DSO-logics are managerially homogeneous in that each has a distinct combination of time-perspective, demand type, process or care pathway configuration, as well as output and outcome –based performance measures. It should be noted that many patient journeys and care pathways might include several of the DSO-logics. Switching between them typically creates discontinuities and coordination problems.

OBJECTIVES
The goal of this project was to develop and disseminate theoretically rich - yet practical - conceptual models, and a toolkit for the development of service production processes. MANAGED OUTCOMES has had the following objectives:
- Development of models of more effective and efficient healthcare systems using a new scientific approach based on Service Operations Management
- Development of tools, methods and models to support the creation of more sustainable healthcare systems that can respond to universal challenges of healthcare demand
- Investigation of relationships between quality of care and costs, efficiency, and accessibility
- Identification of the relationship between healthcare costs and outcomes, using technical, allocative and economic efficiency measures of service production systems
- Identification of different demand segments in healthcare
- Enhancement of cooperation between researchers in Europe to promote integration and excellence of European healthcare systems research
- Development of future scenarios for European healthcare systems

CONSORTIUM
The project brought together a broad group of organizations in a consortium with complementing expertise and abilities to develop these scientific and practical objectives. The consortium comprised the following partners:
- Aalto University, Finland (AALTO)
- Erasmus University Rotterdam, the Netherlands (EUR)
- University of Bamberg, Germany (Universität Bamberg)
- Polytechnic University of Valencia, Spain (UPVLC)
- European Hospital and Healthcare Federation (HOPE), Belgium
- XperidoX, France
- National School of Public Health, Greece (NSPH)
- Balance of Care, UK
- University of Erlangen, Germany (FAU)
- Forum Virium Helsinki, Finland (Forum Virium)

WORKFLOW
The project was divided into three thematic work packages which produced the following outputs. The "Methodological guidelines" (deliverable of work package 2 of the project) provided a detailed description of the research questions and research methodology of the project. This also ensured that the various case studies of the project were conducted in such a way that the results enabled comparative analyses and provided meaningful input for scenario analysis.

The "Report on inventory and analysis of European practises in selected countries" (deliverable of work package 3 of the project) was reporting the results of the research work of the project. It included the review of European healthcare systems, case instance reports on concurrent practices in regions of six European countries, and user experience surveys on patient experiences in regions of six European countries. The combination of the case instance reports with the user experience surveys provided the project with unique insights into the performance of regional healthcare service systems.

The "Report on scenarios of health systems" (deliverable of the work package 4 of the project) described the various future scenarios based on participatory futures workshops with regional stakeholders, and also explored further the operational modelling and financial modelling of regional healthcare service development.

The "Recommendations on healthcare service production systems" was the final summarizing deliverable of the MANAGED OUTCOMES project and thus it is based on several previous deliverables of the project. It presents the main findings of the project and related recommendations.

Project Results:

MANAGED OUTCOMES project work has been based on a structured study of four wide-ranging and richly complex case studies: type 2 diabetes, stroke, hip osteoarthritis and dementia. The primary reason for their selection was that they represent large volume health needs related in particular to ageing and chronic diseases in Europe. The project methodology comprised 1) case instance reports on concurrent practices in each region, 2) user experience surveys on patient experiences, and 3) operational and economical modelling. The project selected to undertake its field research in regions of six European countries, which represent by their financing, service provision and working principles different healthcare systems (Finland, Germany, Greece, The Netherlands, Spain, United Kingdom).

In addition, the MANAGED OUTCOMES project developed and explored the future of regional healthcare through different scenarios about the future. These were based on the analyses undertaken above but were further developed in close collaboration with various regional stakeholders through a series of participatory futures workshops. These developed scenarios assist in identifying potential opportunities for improving health systems and provide policy and decision makers with a range of insights on health system reform.

The key findings of MANAGED OUTCOMES project can be divided in three different areas. There are case-specific, general-level and methodology related findings. The key findings of each of the areas are represented in the following chapters. Furthermore, the project aimed to enhance the cooperation between European healthcare operations management scholars to further promote integration and excellence in European healthcare systems research. It succeeded in bringing together researchers to work closely – both inside and outside the formal scope of the project.

KEY FINDINGS – GENERAL
Regionally organized care pathways and their outcomes suggest that in different management logics, performance measurement and policy objectives should be defined differently with specific focus. The key findings from the selected patient groups can be partially generalised to other diseases with same type of managerial logic.

Chronic diseases with preventable progression, which is represented by type 2 diabetes in the project, have a time-period of months or years and bind the patients to frequent users of healthcare services. These frequent users need to have easy access to services and monitored for balanced care. In addition screening and early identification (i.e. early order-penetration-point) may be key to successful treatment results.

The stroke case represents hyper-acute emergency process and some of the lessons learned can also be taken to other such diseases. The common lessons may include move of order-penetration-point to as early phase as possible, which would allow early recognition of symptoms and rapid access to care. In addition these patients need to be triaged and streamed quickly to the right specialized units for integrated care.

Hip replacement was selected as an elective operation to the study. Even though hip replacement is known as one of the most effective operations it has certainly many similarities with other elective operations. Centralized service provision with high process efficiency allows cost-effective care. With comparison of incidence levels between regions it is possible identify over-use of services and consider different demand management actions.

Dementia care is a fertile ground to study cross-boundary care pathways and how to integrate multiple service providers. Especially, integration of information systems may be a common problem with this type of care pathways.

Taken together the scenarios of different patient groups identify the importance of the regional system in providing the focus for the operations management interventions. This underpins various common features that are likely to be important to future developments of regional health care systems. These include:
- The possibility in EU regions with lower expenditure levels of designing low cost processes that nevertheless generate good outcomes in terms of cost per QALY. These focus on developing processes, often nurse led, that reduce usage of hospital and pharmaceutical resources.
- The relevance of taking a 'whole system' perspective. In all of the case studies we found that important insights were gained from modelling and analysing the combined effects of care processes rather than the isolated effects of individual services or processes.
- The key role of the PHC system to ensure the coordination of services at the operational level across the region. PHC is the focus for much future service development to support care closer to home, patient self-management, and support for carers.
- Major development of information systems is essential to enable different care professionals and organisations to integrate their operations so they can be more responsive and targeted in providing services for patients and carers.

KEY FINDINGS - SELECTED PATIENT GROUPS

1. Type 2 Diabetes
Type 2 diabetes is placing a major burden to the European health systems as it has high costs in care and medication. In addition the prevalence is still expected to grow in the future as the population is ageing and obesity is more common. Clinical guidelines for the care of the diabetes are more or less the same across Europe but regional systems are organized very differently. Therefore it is expected that regional differences in outcomes might be explained by how the processes are organized.

Care of the patients is not homogeneous and division to demand segments helps in studying the provision of the care

Type 2 diabetes is a chronic disease and care usually spans several years or decades. However, depending on the progression of the disease the care needed is different over the years. These different care phases need different types of services which need to be managed differently and therefore studying the patient groups as separate demand segments is helpful. There are several ways to divide the patients to demand segments and there are some good examples which are already in use such as those used in UK. The demand segments used in the project were based on the progression of the disease and needed medication and care. These demand segments made mutually exclusive groups which helped in understanding and studying the disease.

Early diagnosis and prevention of diabetes progression to more severe stages is the key to more efficient care

The Finnish and Spanish regions succeeded to keep patients in early stages of the diabetes and that led to more efficient care. Both regions have had a screening program which probably positively affected to early detection of the patients. The Dutch case instance stands out from the others in terms of incentives and sanctions aimed at GPs which were related to the clinical outcomes.

The amount of care delivered does not necessarily correspond with the best health outcomes

Even though the Dutch region was best on many outcome indicators, they did not provide most care to the patients. Diabetes care is far less expensive in earlier than later stages of the disease. In the later stages when insulin is needed and complications are more common, care is more expensive. The regions which had more focus in early care had better patient reported outcomes and a better care balance.

Easy access to Primary Health Care (PHC) as main coordinator of the care increases health status reported by patient

Care was organized differently in the studied regions. In the Finnish and Dutch case instances this was coordinated by the GP but also heavily supported by practice nurses. On the other hand the Greek case instance utilized more self-monitoring and was led by hospital doctors. Access to the care was more difficult and the patients reported worse health states.

Future scenario: Low cost
This scenario suggests maintaining patients as long as possible in early stages of the condition. It aims to maximise value, expressed as outcome relative to cost, in health systems where overall funding levels are low. There is an important role for generalist nurses in the community to encourage lifestyle change and help to minimise drug use. There is also a key support function for them in doing this from specialist diabetes services usually found in secondary care settings.

Future scenario: Diabetic control
In this scenario active management of patients at all stages of the condition including pre-diagnosis is in the focal point. The level of 'glycosylated hemoglobin' (HbA1c) - and maintaining a balanced level of it - is a key clinical outcome here. Better management requires; 1) information systems to identify and review patients, 2) regular monitoring of patients, with substantial calls on diagnostic services, and 3) the ability to undertake timely corrective actions including medication to maintain symptom control.

Future scenario: Zero complications
'Zero complications' scenario is based on the notion of ensuring that those diagnosed with type 2 diabetes do not experience a 'raised risk' for stroke, AMI (acute myocardial infarction), blindness and sight problems, and peripheral vascular disease and amputations. This requires both enhanced monitoring and treatment of hypertension and cholesterol amongst diabetic patients, and also greater focus on health promotion initiatives. Good registers and information systems are required to support this scenario in order for appropriate treatment points to be recognized and acted on for a given patient.

Regional managerial logic
In the regional organisation of the care for type 2 diabetes the main focus should be in providing easy access to care and maintaining care balance of the patients. Efficiency of service production is in a minor role in this case.

2. Stroke
Stroke is an important challenge for EU healthcare systems. It is one of the most occurring causes of death. It plays a large burden on systems, patients and families. The solution for this can come from a better alignment of services in the regional network of providers. The large differences in outcomes in different countries can be explained to a considerable extent in the way the processes are organized for patients.

Large variations in practices, big differences in health outcomes
There were large variations in the practices and operational measures across the studied regions. One important difference was the way haemorrhagic strokes were handled. In some regions haemorrhagic strokes indicated for neurosurgery were taken care of within the same hospital but in other places patients were transferred to other more specialized hospitals so the specialization level of the relevant SPPs varied considerably. Moreover practices differed in the way the stroke unit was used. In some places all the patients with ischemic strokes went to the stroke unit and in other places only a proportion of all cases was admitted to the stroke unit.

In addition there were extensive differences in operational measures, for instance in the time elapsed from the emergency department (ED) arrival to the patient having a CT scan, length of stay and in the amount of rehabilitation provided. That is, different care pathways had an impact on how smoothly the patient journey could proceed.

These differences seem to lead to considerable differences also in health outcomes. Mortality and patient satisfaction showed relatively large differences and the patient reported health status also varied between the case regions.

Fast speed in hyper-acute phase
Access is essential in stroke care in that the time from incidence to the start of medical care should be minimised. The time from onset to ED door is affected by two factors: recognition of symptoms to call emergency services and speed of the emergency service delivery. As some of the patients greatly benefit from fast access to care it is critical to speed up these two factors. The early recognition can be improved by educating members of the public to recognize stroke symptoms happening in themselves or others. Even though there are large regional differences in distance to the closest emergency department, the difference in travel time is relatively small compared to delays in symptom recognition. From ED door onwards it is all about speeding up the ED process and rapid access to the specialist stroke team, i.e. process efficiency.

Differences in the stroke care processes reflect uneven progress towards Helsingborg declaration goals
The role of the stroke unit varied between the case regions; in some regions the stroke units were the main focus point of care and in other cases the stroke units were embedded in the regional care chain. In the Helsingborg declaration on stroke management with objectives for stroke care, the importance of the stroke service is emphasized. One of the objectives for the year 2015 is that all patients with stroke will have access to a continuum of care from organized stroke units in the acute phase to appropriate rehabilitation and secondary prevention measures. It seems that this objective is fulfilled only in some places in European regions and consequently the best outcomes were reported in cases where all the patients were admitted directly to the stroke unit. In addition unit costs were highest in those places where patients were not always admitted to the stroke unit.

Integration of all the stroke services also varied. The regional collaboration around stroke services ranges between a very light structure with a meeting of partners once a year to a platform for evaluation and development of the stroke services. Therefore some stroke services comprise ambulances, stroke units and medical wards, each working as separate service providers and others are embedded in a regional structure in which all partners in stroke participate (e.g. PHC, ambulance, hospital, rehabilitation, care homes, home care).

Future scenario: Rapid access
Ensuring that patients can access specialist stroke services as soon as possible after the onset of symptoms are reported is in the core of the scenario. Although this allows thrombolysis in cases where the patient is both suitable and diagnosed in time, this scenario prioritises the access to specialist services over the hyperacute phase, as it is appears that the average outcome per patient may be greater, and more productive than further investment in speeding up the hyperacute phase in many cases.

Future scenario: Early identification
According to the scenario improvement in timely access to hyperacute and other specialist treatment requires quicker notification to both the hospital and ambulance services. This, in turn, implies greater awareness amongst the public and, especially those at greatest risk of a stroke. Initiatives within primary health care can support this, and potentially lead on to greater preventive effort in maintaining vascular health.

Regional managerial logic
Organizing stroke care needs to be based on rapid access to the specialist care and fast coordinated care pathway. The service provision points of stroke care need highly specialized resources.

3. Hip Osteoarthritis
Osteoarthritis is the most common type of arthritis, especially among older people and in the hip it can cause pain, stiffness, and severe disability. Hip osteoarthritis can be treated conservative to slow down progression of the disease but only curative treatment is surgery.

Differences in outcomes across the case studies are small and the operation decreases pain and impairments substantially

The outcomes and clinical quality were measured by EQ-5D and reduced pain and impairments. The regions had no significant differences in the outcomes. As the average cost per EQ-5D was approximately 9000 EUROS, the total hip arthroplasties (THA) are very cost-effective operations.

Clinical and process quality of hip replacements is correlated to the volume of replacements carried out

The study results indicate that the productivity and the quality of operations are better in focused hospitals having larger annual volume of operations. In our study, the length of stay was shortest and the share of reoperations and repositions smallest in the unit which had highest annual volume of THAs on both a hospital- and surgeon-level. The hospital with smallest annual volume had the longest length of stay.

The productivity and quality of elective operations can be improved by increased number of repetitions. Many researchers have found an association between volume and performance measured by clinical complications and productivity. The essential drivers are procedure-specific volumes per surgeon and per hospital, not the total volume or size of the hospital. The elective operations should therefore ideally from an efficiency point of view be performed in high-volume specialized centres instead of big multi-specialty hospitals.

In terms of productivity, length of stay and operating room scheduling are the key factors. The length of stay affects significantly the total costs of operation as the ward care is very resource intensive. In operating rooms the differences between throughput times (surgery or Patient-In-OR-time) are minor, so the differences in operating room productivity between hospitals are due to differences in scheduling and in utilization rates.

In terms of operations management, high quality and productivity is achieved by focusing on
1) waiting list management,
2) sophisticated operating room scheduling
3) reducing length of stay by pre-defined care pathways

Waiting times are a common problem in EU countries but there are successful examples of reducing them

Three out of five regions had over 200 days' waiting time for the operation. In those regions, the waiting time to specialist assessment varied from two to three months and the waiting time from surgery decision to the operation from five to six months. In two regions, the pressure for reducing the waiting times had increased and some successful methods have been applied to reduce the waiting times.

The Dutch case instance region had on average 60 days waiting time for the operation. One reason for reduced waiting time is successful competition between hospitals. In addition, the orthopaedics department of the studied hospital participated in 2004 in a national quality improvement programme 'Faster Better'. The objective of this programme was to reduce access time substantially to a maximum of one week. The hospital is active in applying 'Lean' management methods, and new scheduling methods have been developed as well as a reduction in waiting times.

In the UK region, the average waiting time for the operation was 100 days. There have been major initiatives in the UK in recent years to reduce waiting lists for hip and knee replacements, most attempting to achieve this by increased funding to encourage additional operating sessions within existing resources, or using additional resources from the private sector. The studied elective orthopaedic centre was opened in 2004. It was set up by an alliance of Primary Care Trusts (service commissioners) and hospitals in response to a major shortfall in the number of hip and knee replacements being undertaken, which was leading to very long waiting lists. The centre is focusing to the total joint replacement operations having an annual volume of over 2000 operations.

In Finnish case instance region, the waiting times have been reduced after the national care guarantee was applied in 2005. But as the care guarantee allows three months waiting for specialists visit and six months waiting for the operation it does not cause any additional pressure to reduce the waiting times. Thereby the actual waiting times are quite near to the maximum allowed limit.

In summary, the successful solutions for decreasing the waiting times has been 1) increased competition between hospitals (Dutch case instance) or 2) increased capacity combined to a focused hospital (UK case instance). If demand management is not combined with those solutions, the increased competition and/or capacity in the region can lead to increased number of operations instead of reduced waiting time.

Regional costs of hip-OA are dominated by the incidence of hip replacements and not by process efficiency

The differences in incidences had significant influence on the total regional costs of hip replacement services in a region. Since the differences in incidences are counted as multiples and the differences in unit costs are only 20-40 % between hospitals, the key question in terms of the regional costs of service delivery is the incidence of operations – not the productivity of processes.

The study indicates that the incidence of operations partly relate to the availability of resources. In publicly funded or insurance-based system the marginal rate of return does not limit the demand, so demand management is the key question in terms of maximizing the regional cost-effectiveness.

Future scenario: Process quality
The key aspect of the scenario is that patients are treated as quickly as possible, in terms both of the wait for treatment and the length of stay in hospital. The main requirements for improved process quality are 1) good information systems to minimise waiting times and schedule resources, 2) flexible theatre session times, 3) advanced nurse practitioners to manage patients into theatre and through the recovery process, and 4) adequate volume of procedures to ensure expertise maintained across the different clinicians.

Future scenario: Managed demand
Closely tied processes arranged between primary health care and the elective hospital services to manage the thresholds at which patients are treated. This could include interventions at younger ages, but with conservative treatment as the norm. The Managed Demand scenario would require:

- Pathways agreed by GPs and orthopaedic surgeons across the regional system regarding operating thresholds, including a pain threshold element
- Clinical decision support systems to support and control patient referrals
- The introduction of less invasive procedures for patients below these thresholds.

Regional managerial logic
Process efficiency and demand management are the key aspects in organisation of hip osteoarthritis care and respectively elective operation for it. As there are no specific need for easy or fast access the care can be centralized to high volume hospitals.

4. Dementia
Dementia is not a single disease, but a non-specific illness syndrome, which is an increasing issue for both care management and resource usage for older populations across the EU. The services required can be costly and complex to organise over a long time period. There is a large burden on health care systems, but especially for patients and their families.

Complex care network needs integration and coordination to support carers

In all of the case regions dementia care was provided by multiple stakeholders. These stakeholders form cross-boundary care pathways across professions, disciplines, systems and organizations. To provide seamless care to patients and support informal carers in the care process all the stakeholders need to have a shared understanding. This is best achieved through integration of components, such as perspectives, skills, and information, to create a system. Each component needs to contribute, as well as adapt to the other components thus fusing various contributions to something new. In continuous care integration should produce an understanding of the situation shared by all relevant stakeholders, and support the development and execution of a corresponding care plan.

In healthcare formulating a diagnosis is a typical result of integration. Various informational contributions from medical history, examinations, and assessments are brought together and merged into a shared understanding of the situation. The contributors of various informational inputs need to engage in "negotiation of meaning", as it is not always obvious that various parties have a common view or identical objectives. An integrated view of a patient's situation should result in a care plan that is executed in a coordinated way. The final integration takes place in the patient's life-world and is measurable as outcomes. Obviously, an integrated diagnosis and a coordinated care plan will not remain intact endlessly. If new inputs appear, integration needs to be repeated in a cycle of continuous adjustment and learning.

Memory clinics lead to earlier diagnosis in the Finnish and German cases
In the Finnish and German case instances memory clinics were a part of the care pathway for the dementia patients. Even though memory clinics are widely suggested as a good practice in the literature there were few or none in the other case study regions. They are expected to bring together professionals with a range of skills for the benefit of patients, carers, and colleagues.

In both of the regions with memory clinics patients were diagnosed on earlier at less severe stages than in the other case study regions. However, it was not clear that this led to better outcomes for patients (as in other case studies) since there were no identified care processes available for such patients.

Earlier admission to care homes in the Finnish case; community-based care for all people with dementia in Greek case; other countries in between
One consequence of early detection of dementia in the Finnish case instance is that patients are institutionalized on early stages of the care. This may be affected by cultural aspects; Finland is individualized country and there are not such family communities as in Greece. However, this may also be due to features of service supply; in contrast to the Finnish care model, there were no care homes in the Greek case study region. People with dementia here are cared for at home until very late stages of the condition and the organization of the care is community-based.

The care models in the other case study regions were positioned between Finnish and Greek models.

Total hours of care (including formal and informal) are more in 'south' than 'north'

In southern countries (Spain and Greece) more time was allocated for caring of the patient than in the northern countries. The differences in levels of support also help explain the differences in discharge outcomes from the acute hospitals identified in the case study. The Finnish practice is to discharge to a step down facility, whereas in the Greek case instance patients were nearly always discharged home. There is a close correlation between the total hours of care available to care for the patient at home and the probability that a patient will go home immediately after an acute hospital episode.

Future scenario: Full integration
The scenario includes integration of all processes that can be identified to ensure that people with dementia can continue to live dignified lives, including remaining in their own homes for as long as possible subject to the quality of life for them and their carers. This requires; 1) good information systems to allow shared care across different agencies and services, 2) close involvement and support to carers to enable them to maintain their caring role, 3) a full range of services to meet direct care needs and those arising from comorbidities, and 4) memory clinics to provide early diagnosis (assuming preventive therapies become available in the future).

This represents a full range of services that would bring about integrated care for people with dementia.

Future scenario: Hospital coordination
The Hospital Coordination scenario recognises that development of integrated care will be uneven, and it may be more productive to introduce processes in those services that may be best able to lead other services and service providers. This scenario would require:
1) assessment on admission of older patients for memory and cognitive problems,
2) liaison and outreach nurses to link to other services and support inpatient services,
3) data sharing arrangements with other key services, and
4) carer support services.

Future scenario: PHC coordination
The PHC Coordination scenario aims to exploit processes being established more widely in PHC, and thus would provide a way to progress in regions where PHC is being prioritised. This requires:
1) opportunistic or routine assessment of older people by GPs to identify dementia,
2) use of information systems to ensure services are notified of need and can track care histories,
3) community liaison staff (nurses or social care) to coordinate and support care delivery, and provide in-reach to acute hospitals.

The PHC Coordination scenario is similar to the Hospital Coordination scenario in its focus on opportunistic diagnosis and the use of liaison roles and information sharing.

Regional managerial logic
The dementia care is based on a multi-provider network. In effective care organization it is critical that these providers work is coordinated and integrated.

KEY FINDINGS – METHODOLOGY
MANAGED OUTCOMES is based on three methodological principles: a split between output and outcome, regional service systems as the unit of analysis, and segmentation along demand-supply-based operating logic. Using these principles the MANAGED OUTCOMES approach to regional health care delivery systems for specific patient groups was elaborated, consisting of the following components:
1) An operational model for describing the relationship between demand-services-patient journey-resources-costs. This serves as the backbone of the MANAGED OUTCOMES approach.
2) A provider and user perspective on the performance of the regional delivery system, using data from information systems of provider organisations and experiences of patients from a survey held under the users of the regional health service delivery system.
3) A comparative analysis that compares the performance of regional delivery systems for the same patient group in different countries or settings, to learn from the differences in health care practice, and to identify best practices.

And furthermore at meta level:
4) An economic evaluation of outcomes of the health service delivery system related to the costs of inputs/resources
5) Scenarios for the future development of health care demand and health care services
This approach was tested for four cases: type-2 diabetes, stroke, hip osteoarthritis and dementia.

The results of using this methodology for regional health service delivery systems for specific patient groups can be summarized as follows:
1. The MANAGED OUTCOMES approach to regional health service delivery system was very successful in developing a multi-perspective and rich picture on the functioning and performance of regional service delivery systems for specific patient groups. To have an operational model of a healthcare delivery system and combine that with data from information systems and user surveys gives a wealth of information, and MANAGED OUTCOMES has proved to be able to handle this amount of information in a manageable way.
2. The comparative analysis allowed us to compare different practices across Europe and identify potential good practices that offer possibilities for learning and improving services, despite the differences in healthcare systems and cultures.
3. Though the cases investigated were first of all meant to elaborate the generic approach of MANAGED OUTCOMES for a specific patient group, the results achieved for the patient groups investigated look very promising. While the limited amount of cases does not allow for statistical testing, the multi-perspective approach resulted in many interesting conclusions to improve healthcare practice by developing more effective and efficient health service delivery systems.
4. The meta level economic evaluation of the outcomes of services in relation to the costs of services, and the development of scenarios turned out to be important components in the MANAGED OUTCOMES approach that worked out well in combination with the insights from the operational model, the operational performance and the user evaluation of health service delivery.
5. The components of the MANAGED OUTCOMES approach at the operational level of services worked better for type 2 diabetes, stroke and hip osteoarthritis than it did for dementia. This is understandable as the dementia case is very different from the other three cases, and did not allow for an operational model along the lines of the other cases.

Potential Impact:

The impact of the MANAGED OUTCOMES project can be seen in several areas. The impact of the project is in the following described using two different approaches:
a. impact of the new concepts and knowledge created by the project and
b. impact for the various stakeholders of the project.

Impact of the new concepts and knowledge
The MANAGED OUTCOMES project was successful in creating an initial understanding on how healthcare outcomes and costs are affected by the efficiency of service production, the time and location constraints within a regional healthcare delivery structure and the degree to which people are empowered to participate in the co-production of their care. With the particular focus on regions as essential entities in healthcare service operations, the project also increased the understanding of the regions and their role in health services. The health policy research and debate has so far been focused on national systems, and this research shows significant differences between regions. Thus the work of the project has visualized that regions are meaningful units of analysis for health system research.

The work of the project has pointed out that increased attention is required to naturally developed regions in the development of healthcare service operations. This research work shows that regions as essential entities in healthcare service operations. The research also showed that the construction of optimal service healthcare delivery systems, the focus should be on naturally developed regions. The particular role of the regions has not been sufficiently discussed and explored in the development of healthcare service development. This project showed also that one key challenge of the European health systems lies in the effectiveness and efficiency of regional health service delivery networks for specific patient groups.

The research work of the project highlighted that the focus should be directed to health outcomes instead of output volumes and expenditure. While outcome measurement can still be laborious and not always precise, it can be successfully used to assess the performance of care pathways. The project succeed in its key objectives to identify of the relationship between healthcare costs and outcomes, using technical, allocative and economic efficiency measures of service production systems. Also conceptually, it was important to deepen the understanding and identification of different demand segments in healthcare.

The research work undertaken in the project illustrated that health services need to be segmented according to their operating logic. The point-of-departure in the MANAGED OUTCOMES project was that healthcare is a cluster of industries, each with its own operating and economic logic. Based on this principle, the Demand-Supply-based Operating logic (DSO) was developed as the segmentation principle. The DSO model is based on the assumption that demand and supply conditions together require that service production takes on a specific operating logic. Although the four cases investigated in detail in the project do not yet constitute a validation of the DSO-logic model, it has proven to be useful in analysing the differences between different care pathways. Therefore objectives and performance measures should be developed to reflect the different operating logics.

One elementary finding of the MANAGED OUTCOMES project was that from a total cost perspective, process efficiency improvement has less impact than demand management. The focal aspect for the development of regional health systems should thus be rather on demand / incidence management, which can contribute more on the total cost, than pure process efficiency development. The results of the MANAGED OUTCOMES project show that the Order Penetration Point (OPP) and demand can be actively managed. Moving the OPP earlier is advantageous in some cases (i.e. type 2 diabetes), while it can increase demand and lead to less cost-effective procedures in some others (i.e. hip osteoarthritis).

The work of the MANAGED OUTCOMES project also emphasized that in researching healthcare operations, it is important to include both clinical and statistical data as well as data on user experiences. The project succeeded in combining in its analytical work these complementary aspects. Furthermore, the work with the scenarios meant that the project in its research was not only looking back, but also looking forward using participative working forms and thus also assisting the regional health care stakeholders to build new future approaches for regional health care. Our work visualized that the development of regional healthcare service delivery networks requires innovative multidisciplinary approach. The MANAGED OUTCOMES project developed an innovative multidisciplinary approach to investigate regional healthcare delivery systems. The methodology adopted helped to use the combined perspectives of operational model, operational performance data on quality and service, user experiences, economic evaluation and futures literacy.

The work of the project was based and deeply rooted in various regions in six European countries. The working approach of the project relied much on the regions from data collection to future scenario development. The recommendations the project made based on its work were directed to improve the regional health service delivery systems.

An important factor in the impact of the MANAGED OUTCOMES project is directing the attention for the development of future scenarios (such as low cost, diabetic control, zero complications) with explicit outcome and cost projections in order to better articulate policy choices. Since there is no optimal solution to the service distribution trilemma, outcome-based objectives exhibit significant variation, and resources differ across regions and care pathways, there is a need to develop and articulate different scenarios for policy makers and voters to choose between. This project successfully developed the 'Futures Literacy' methodology for obtaining views of stakeholders which could then be combined with other data sources to produce well-grounded scenarios.

The successful work of the project has been based on the effective and seamless collaboration between its partners. This collaboration and shared mutual interest has been the basis for high-quality results of the project. Also feedback from stakeholders underlines that the project has generated vast amounts of interest from many different groups and that work has been relevant and is appreciated.

The fluent research work was based on the thorough development of methodological approaches, definitions and guidelines during the early stages of the project, which were punctually documented in the methodological guidelines of the project. This experience highlighted the importance of precise planning in joint European research projects and assisted greatly in the actual research work undertaken by the project partners.

The MANAGED OUTCOMES project produced well-researched written reports, which serve as the basis for articles in scientific journals as well as other dissemination activities. In addition, the MANAGED OUTCOMES project used also novel dissemination approaches – e.g. webcasts of the final conference – to spread its message and findings to wider audiences.

Impact for the various stakeholders of the project

Another, complementary view of potential impact is to approach potential impact according to the various stakeholders of the MANAGED OUTCOMES project. Within the work of the project, we developed a stakeholder map and we also discussed the relative power and interest of the various stakeholders according to the work and development of regional healthcare systems. The key identified stakeholders were:
- patients (including their families, relatives etc.)
- healthcare professionals
- healthcare service provision points (GPs, clinics, hospitals etc.)
- regional authorities and decision makers
- universities and research institutes
- national healthcare decision makers (ministries, governments, parliaments)
- national associations, networks and interest groups (e.g. patient associations, trade unions)
- companies (e.g. pharmaceutical and medical device manufacturers)
- European and global healthcare decision makers (European Commission, European Parliament, WHO etc.).

- Associations, Networks, Lobby and Interest Groups
(i.e. international, European, national, federal, regional, local)
Industry (i.e. pharmaceutical, medical, technical)

Ministries, Policy and Decision Makers Families, relatives and friends Key health care professionals Employers, companies,
trade unions
Patient Research
Community
Service Provision Points (SPPs)

Insurance Companies
European Commission

The potential impact for the patients and also their families and representatives (e.g. local patient associations and other local third sector actors) is the opportunity to compare in the four researched cases the various healthcare practices and processes in various regions of Europe. One important part in modern healthcare systems is the empowerment of patients, and thus the openly available research results of various healthcare practices are also providing the patients more resources in their dialogue with the healthcare professionals. An elementary part of the research work of the MANAGED OUTCOMES project was the collection and analysis of the user experiences, and thus this approach also highlighted the importance to bring the users and their service perception as an important element to the assessment and evaluation of healthcare service delivery systems.

Furthermore, in the Futures Literacy workshops the MANAGED OUTCOMES project involved also the various regional patient associations as important stakeholders to the work and thus presenting the voice of the patients in sketching the future of the regional healthcare systems.

The potential impact for the healthcare professionals and healthcare service provision points stems from the comparison of various practices in different regions. The project could document and visualize appropriate good practices, which can also serve as benchmarks in the development of professional practices for healthcare professionals.

The research results of the MANAGED OUTCOMES project confirmed that coordinated early interventions are reducing healthcare costs and improving patient experiences. Our case findings showed the importance of coordinated actions between various actors in regional healthcare service provision, and that complex care networks need integration and coordination to support carers. As it is well-known in the research literature that early interventions e.g. in diabetes-2 and dementia can be both cost-effective as well as increase the quality of life of the patients. Our research confirmed that early diagnosis and prevention of progression to more severe stages is the key to more efficient care as well as to increased patient satisfaction. Thus the coordination of the work of various healthcare professionals is essential.

Making an impact – dissemination activities of the project
The MANAGED OUTCOMES project consortium was active in the dissemination of the results Dissemination task workforce was founded in the beginning of the project to design the publication strategy and identified the most valuable dissemination components. This task force defined a first strategic plan that was summarized in the Dissemination Guidelines. Subsequent meetings of the Dissemination Task Force were made to refine and update the dissemination plan. The task force was responsible for creating official dissemination materials, updated according to the progress of the research and presenting the main project results, including: leaflet, summary, presentation, certificates.

List of Websites:

http://www.managedoutcomes.eu
142630691-8_en.zip