Integrated care is the collaboration of multiple services around a client with multiple needs. Many problems with quality occur as a result of failures in communication and co-ordination of services. Put another way, improvements in one link of the chain may be neutralised by a failure in some other link. If the co-ordination and collaboration between the relevant professionals and organisations takes place only on ad hoc initiatives, there are various potential factors that can prevent a high-quality, cost-effective care from being achieved. If, on the other hand, the various actors work as a network or a team, with shared goals and protocols, then the integration is better structured, and there is a better chance of reaching the quality objectives (see also Chapter 2).
According to the Saskatchewan Commission on Medicare (2001), ‘Health care organisations that systematically emphasise quality are the best places to work. They respect and maximise the contributions of all staff, they reduce the amount of unnecessary and ineffective work, they reduce error rates and they produce better outcomes and job satisfaction.’ There is no reason to assume that this would not apply also to social welfare services.
As yet, there is little experience or literature on quality management for integrated care in Europe, but, taking the nature of integrated care into account, it must also be based on collaboration and information exchange. Working in an integrated care system certainly makes quality management more complicated, but there are models that can be used, and lessons that can be learned, from the research and practice of generic quality-management approaches.
There is a wealth of practical experience demonstrating that systematic quality management has increased people’s motivation in relation to quality and awareness of costs, thus leading to improved performance. Experience has also shown that improvements in quality can be achieved with small changes in attitudes and daily routines (Olsson 2003). Research points out that 85 per cent of health care quality is determined by the system performance. In other words, quality is a result of teamwork rather than of an individual profession (see Medical College of Georgia, Web links).
This chapter is about these models, approaches and lessons. The key message is that quality management in integrated care is a continuous learning process for all participating in it. The arguments do not directly depart from any special model or certain terminology and, due to the lack of models on quality management for integrated care, it draws heavily on generic quality management approaches. This illustrates the first lesson about introducing a quality management system in integrated care: accept that there are several approaches and terminologies for quality and that this diversity can be overcome only by common discussion where diverse professional cultures are respected as equally important.
The objective of quality management is to support the provision of best possible care outcomes and efficient use of resources. It is expected to facilitate:
The key terms relating to quality of care are inputs/structures, process and outcomes (Donabedian 1980). The CARMEN Network (Vaarama and Pieper eds 2004) defined the quality of integrated care of older persons as involving the following elements:
Sources: Evers et al (1997), Øvretveit (2001), Vaarama and Pieper (2004b)
In integrated care, the quality of a multiple service can be evaluated only in terms of looking at a chain of actions taken by different organisations and professionals that are co-operating or transferring responsibility from one to another – for example, from hospital to home. So the mission statement, outcome objectives and measures, processes, and protocols must be agreed by all actors participating in the care network.
Clarifying the objectives is the starting point for defining right outcome measures. To be able to say what sort of outcomes should be evaluated, one has to ask questions such as:
This activity should be organised as an iterative and interactive process involving everyone participating in the care chain. Multi-agency structures and protocols are needed to add coherence to the objectives. This collaboration should also involve clients and their carers, and should empower them with a meaningful participation. That way, the quality management process becomes a dynamic learning process during which the quality of care improves step by step. Building up co-operation for quality between different professionals, units and organisations serving the same clients is the starting point for implementing any quality scheme in an integrated system. Fig 11 illustrates this learning process.
Quality management can be divided into three separate levels:
These are explained below:
At this level, the objectives of the care of an individual client are set, the interventions are customised to meet individual needs and circumstances and documented in the care plans, and the delivery and processes are recorded. The quality management questions at this level may be:
The client’s needs are not static, but change over time, so a regular monitoring and evaluation is needed to make the necessary adaptations in time. (See also Chapter 4 and Chapter 6).
At this level, the quality management questions concern issues such as co-ordinating the delivery of the multiple services to the client, the collaboration of the professionals, ensuring unity within the objectives and goals of frontline professionals, and the degree to which professional standards of care in practice are met. Two topics especially important to monitor at this level are the simultaneous dimension (contents of care and logistics) and sequential dimension (care following the needs of the clients) of integrated care. (See Chapter 1).
At this level, the quality management questions deal with issues such as access to care, financial sustainability of the care system and equity of distribution – in other words, who loses and who gains, how equitable the distribution has been in relation to the client’s needs or needy groups, how great these needs are, and what types of unmet needs can be identified. System-level quality management is expected to increase accountability and to provide information for evidence-based decision making.
There is a range of models and approaches, of which total quality management is the leading option for integrated care. Regarding the evaluation of the quality of care, there is a range of measures available, but none is tailored for integrated care. Measurement of cost-effectiveness of integrated care is especially demanding.
Total quality management (TQM) is currently the most common quality management strategy, and the existing models and approaches are more or less applications of that strategy. TQM is a management philosophy central to the organisation’s goals, rather than a tangential activity. The goal is to produce an organisation-wide plan with specific quality goals. This involves work being done to get all personnel committed and enthusiastic about quality – all the way down from top management to the ‘floor’ level.
TQM aims to cross departmental and disciplinary boundaries. When it is first applied, it requires a substantial investment in training, and involves an ongoing process of quality improvement rather than a one-off setting of quality standards. TQM emphasises the prevention of mistakes and defects before they occur, rather than correcting them retrospectively (adapted from Pollit 1997. See also Pieper 2004 and the ‘quality’ website, Web links).
Quality award schemes are not quality systems as such, but they do offer a basis on which a scheme can be built, and are increasingly used to build up quality management systems for care organisations. One example is the European Quality Award. There are also many national awards available (see Øvretveit 2001).
Source: American Hospital Corporation (1989)
National or regional governmental agencies inspect or accredit care organisations to see whether they meet the defined standards. One example was the UK Commission of Health Improvement, which audited services every four years or whenever a serious quality problem was suspected. Nowadays, state initiatives, too, have moved away from control, moving instead towards stimulating quality development, as can be seen from the Finnish and Dutch examples presented later in this chapter.
This is where governmental or non-governmental accreditation agencies award a ‘credit’ to a service to indicate that it has met defined standards. The best-known scheme in health care is the US Joint Commission for the Accreditation of Healthcare Organisations (see Web links).
ISO 9000 is a set of generic international quality standards. The scheme certifies whether an organisation meets standards for a quality management system. Under the ISO 9000 approach, an organisation seeks registration for particular processes or systems. It pays a fee and, after a period of preparation, the chosen process or system is assessed by an approved, external, third-party assessor. If the assessor decides that the process meets the requirement of the standard, then the organisation can display its accreditation on its products and literature. The ISO 9000 has been used most often in laboratory and non-clinical support services.
Peer review concentrates on whether providers meet clinical or other standards of care. Here, the work of an individual professional or group of professionals is assessed by a group of fellow professionals. The review can vary from systematic assessment with defined standards to more informal peer review visits, where the assessments are offered in a supportive way, with a view to encouraging improvement. The peer review may be internal or external to the organisation under review, it can be voluntary or mandatory, and the frequency of the reviews can vary greatly. There may or may not be sanctions in use, and the outcomes of the review may be confidential, or may be freely available within the public domain. For example, the UK National Health Service Medical Audit is a form of internal peer review, while the US Medicare Program Peer Review is external.
The basic idea of benchmarking is straightforward. First, an organisation identifies one or more of its processes where it wishes to make quality improvements. It then looks for one or more other organisations that appear to have achieved high degrees of excellence in that same process, and designs and implements a plan for raising its own performance up to or beyond the level of that of the benchmark.
The Netherlands government uses benchmarking to stimulate quality and efficiency of care, and it has developed performance indicators for home care providers, in consultation with them. The dimensions measured were: efficiency in terms of costs and outputs, quality of care as evaluated by clients and staff, and stakeholders’ appreciation. It constructed a combined score to identify ‘best practise organisations’. This enabled organisations to compare their own performance to general performance, and to the performance of organisations providing similar types of care.
Analyses of the dimensions allow indicators, as factors that contribute to the specific value on that dimension, to be traced. With relevant refinements and adaptations, benchmarking has also been applied for nursing homes. One of these refinements has been that the indicators will be clustered in dimensions according to the INK management model, the Dutch counterpart of the European Organisation for Quality (EOQ). (This management model is derived from – and is quite similar to – the European Foundation for Quality Management – EFQM). This will allow better insight for managers on how the results are derived, and how to improve their management.
Source: Huijbers (2003)
This collaborative model was developed in Sweden in palliative care, but it can also be implemented in other fields of care. Olsson (2003) describes the model with the phrase ‘small changes – great improvements’. The model does not require any major surveys or drawn-out decision processes to get started. Cross-professional teams from different organisations and authorities are drawn from personnel working at the hub of the profession, who must ask themselves the following three questions:
The changes are measured regularly from the time that a decision has been taken on what is to be improved and what should be measured.
The model project has provided useful lessons for improving the quality of palliative care. The most important of these has been that great improvements can be made using very small means and very quickly. In one example, a teaspoon of turnip rape oil every day eliminated both anxiety and constipation in dementia patients at a nursing home. In another, regular pain checks using the ‘VAS’ scale resulted in reduced silent suffering among palliative patients in a nursing home. In a third case, writing down stories from the lives of clients when they were welcomed into the nursing home was used to create care based on individual needs and requirements.
The first breakthrough series took nine months and the second took six. During the process, many teams members bonded, and the collaboration has provided a starting point for improved co-operation in other, quite different, areas. This Swedish group is now planning a breakthrough series on dementia care.
This is an internationally validated, comprehensive assessment instrument for long-term care of older people. It offers tools and processes for:
The results can be computed at the individual client level, the organisation or team level, and on higher aggregated levels.
RAI was originally developed in the United States as a response to criticisms about the quality of long-term care of older people. In order to use it, one must apply for a licence from InterRAI. InterRAI is an organisation of the developers of the RAI-instrument, and it owns the copyright of it (see InterRAI, Web links). At present the RAI-instrument is already validated – or in the process of being validated – in many European countries, and is likely to be one of the most-used assessment and evaluation instruments in the long-term care of older people in the future.
This software application is a strategic macro-level quality-management system for the social and heath care of older persons. It includes key indicators of need, supply, staffing, costs, effectiveness of goal attainment, labour productivity, economy of care, and equity in terms of access to services and unmet need. The application offers two modelling options (‘trend extrapolation’ and ‘planning by targets’) for systems improvements, and enables the user to simulate the costs of alternative options.
The software, which is maintained by the Finnish National Research and Development Centre for Welfare and Health (STAKES), makes use of existing care registers. If these are not available, it is possible to build up the database by collecting the data separately. The results are given in popularised tables and graphics and designed to be understood by lay people, rather than only specialist planning experts.
This system facilitates fact-based decision-making and planning. Supported by a collaborative planning model, it brings the social and health care planners, professionals and other stakeholders together to plan, monitor and evaluate the care of older persons, to identify the needs for improvements at the system level, and to compare alternatives to be able to select the most promising one.
The first version (‘EverGreen’) was introduced in 1990, the second (‘EverGreen 2000’) in 1997 and the newest version (‘EverGreen 2000 Plus’) in 2004 (see Vaarama 1995, 1997, 2004). An application in English will be available in 2005.
Monitoring and evaluation of quality of care are the core tasks of quality management.
To be able to monitor and evaluate, one must be able to measure quality outcomes and costs. The key terms of quality measurement are quality criteria, quality standards and quality indicator:
There is a common desire among care managers to be able to measure the care outcomes in terms of cost-effectiveness, or – as care practitioners often say – ‘cost-benefits’ of care. By definition, the cost-effectiveness of care is the ratio between the costs and effects of a given care service. A service may be effective in meeting the needs of a client, but it is not cost-effective if similar outcomes could be achieved at less cost, or if better outcomes could be achieved for the same cost. This is very important information for the care managers, but there is still a lack of valid indicators that would be easy to administrate in practice.
The problem deals with measuring outcomes and costs alike and highlights the importance of choosing the right outcome measures. It might be extremely difficult to find out the only outcome measure, but a good alternative is to look at the separate dimensions of care outcomes. For example, it may not be possible to make valid measurements of the improvements in the client’s quality of life that are really caused by the integrated care they receive. However, it is possible to use measures such as reduced pain, or reduced emotional or behavioural disturbance in the client or reduced burden on the principal carer, and it is always possible to ask the client about their levels of satisfaction and their subjective levels of health, mood or well being. There is a wealth of validated scales and tools available for measuring these dimensions. (See, for example, the Growing Older Programme, detailed in Dean 2003).
In addition to quantitative measures, qualitative information is necessary for a comprehensive evaluation of the care quality. For example, many process objectives are qualitative, and relational aspects of care (face-to-face care) must be measured qualitatively. However, when qualitative objectives are clearly defined, they can often be measured by using quantitative scales. In integrated care, defining these objectives and agreeing on respective indicators is one element of the quality management cycle described earlier.
As said earlier, it is better to take multiple outcome measures than taking just one, as this leads to more a comprehensive evaluation and increases the validity of the evaluation. This is especially true in integrated care, where several objectives for quantity, types and quality of care are set for one case, and several professionals – and usually several organisations too – are involved in the care. Evaluating the quality by looking at only some of the activities that take place would not be sufficient. So it is better to have a number of success indicators than just one. It is also desirable to have a variety of different perspectives on the same outcome. For example, in the UK Duffy et al (2001) have evaluated service quality in nursing homes from the point of view of residents, family and administrators.
However, having a number of outcome indicators for the same case can cause problems for cost-effectiveness evaluations. So, for cost-effectiveness evaluation, a primary outcome should be defined against which the costs can be measured and the cost-effectiveness evaluated. All the costs caused by the multiple services need to be summed up and related to the primary outcome. The primary outcome should be measured in units that enable a cost-effectiveness ratio to be calculated. It is well known how difficult cost-effectiveness is to measure, but this does not mean that it is not possible. This topic is little discussed in integrated care, so there are no models that can be presented here. Managers interested in this topic should look at the existing health-economic literature. For information about measuring cost-effectiveness of social welfare services, see Sefton et al (2003).
The practical implementation of quality management schemes can be separated into four steps that follow each other or are implemented simultaneously.
Deciding which model to choose depends on the problem to be solved and the circumstances in which it is to be implemented. Planning carefully how to introduce a quality management system may be more important than your choice of model.
Integrated care is all about co-operation within and between different actors, professions, roles, organisations and other systems. Cultural and professional differences between these various elements are significant. To ensure ownership of a common quality effort, everyone involved must have the chance to discuss and decide which concepts and goals they will adopt, how to fit them to their own work, and what are the basic values of their work (in other words, a mission statement). Step by step, the following activities may lead to a commonly shared quality system.
Quality improvement is a continuous learning process, and in an integrated system it must be carefully organised. The organisation needs to provide a clear framework for shared quality management, and must ensure the managers and personnel feel motivated to continuous training and self-education.
It is crucial to have right-outcome measures. To choose these measures, organise a process in which all organisations and key professionals are involved.
For further information relating to Step 4, see also Sefton et al (2003).
Although monitoring and evaluating quality of care are crucial elements of quality management, in fact evaluation of the efficiency of diverse quality management systems is sparse. At the minimum, a ‘before quality management versus after quality management’ evaluation should be employed to evaluate whether the new system is actually improving results. ‘Before’ evaluation should be performed before the quality management system is introduced to the team. ‘After’ evaluation can be done in conjunction with monitoring and evaluation of care outcomes and costs. Monitoring and evaluation may need to be repeated several times before drawing crucial consequences – it takes time and effort to change!
Unclear goals, insufficient resources and bad timing can hinder the entire process, reduce motivation and lead to poor results.
Another barrier to overcome is poor-quality documentation of care processes and interventions, and lack of commitment of the staff involved. Documentation of care varies considerably even within one organisation, let alone among many. This barrier has to be overcome at the initial stage of the quality management system being brought into practice.
There are also a number of potential barriers to effective collaboration between professionals. These may include:
In addition, there may be parallel policies or quality schemes implemented by the professionals and organisations participating in the integrated care network. These may compete with the common effort to be taken, and affect the legitimacy of the planned process.
Care must be also taken when introducing new models, visions and language to a working team. Welfare and health professionals may regard models taken from the commercial sector with scepticism, or even cynicism. They may also have fears about possible criticism arising from the evaluation results – especially when clients are empowered to express their opinions.
The introduction and implementation of integrated quality management can be proactively supported in many ways, and the means need not to be major and costly. To prevent barriers to effective collaboration between professionals, it is usually possible to achieve reasonable common understanding by organising enough time for group discussions about goals and results. At first glance, the aims and terminology of different organisations, or of individual professionals may appear to vary considerably, but after getting to know one another better, significant overlaps may become apparent (see also Chapter 9: Cultural change). A crucial and more expensive support is to produce a common documentation system that all the parties can use (see also Chapter 12).
In the case of parallel policies or quality schemes within the integrated network, or in the separate organisations, it is important to set out a clear contract on what policy the integrated network will follow. All professionals and organisations involved in the integrated care network need to commit to common goals, procedures, responsibilities and checkpoints, and investigate whether there are contradicting goals in some other implemented procedures. If this is the case, the network must create a consensus that bridges the old and new procedures, to ensure that the efforts towards quality improvement can be carried out as intended. These issues must be contracted in a written protocol that everyone involved can access.
Finally, for practical success, it is important that the quality management initiative has the highest level of commitment, and that it is accepted by the highest decision-making body. It is also vital that the initiative is guaranteed adequate resources, such as time and information systems, before starting.
It is a great help if there is an official policy for quality improvement accepted for the country, region or organisation in question. For example, the Finnish government has worked with relevant stakeholders to develop the Finnish National Framework for High Quality Care of Older Persons (Ministry of Social Affairs and Health/Finnish Association of Local Authorities 2001, Vaarama et al 2001). These national recommendations emphasise integrated needs assessment, evidence-based interventions, timely service, seamless care chains, humanity and dignity of care processes, sufficient and qualified staff, and co-ordination, collaboration and co-operation in planning, delivery, monitoring and evaluation of care. The government has also earmarked some finances to facilitate realisation of the recommendations. To support implementation, the ministry contracted with STAKES to provide a set of national performance indicators, which are available online (STAKES 2003).
Strategic tools such as ‘balanced score cards’ (Kaplan and Norton 1996) can support quality management considerably, as the experiences of the city of Helsinki tell (Valvanne 2004), but in integrated care this may apply only provided that all participating organisations share the same strategy. Also very helpful are evidence-based professional standards, or such practical results as, for example, the breakthrough model provided for palliative care (Olsson 2003).
The discussion about the management of quality of integrated care is sparse, and so are the models, methods and measures for it. This chapter emphasises that the quality management of integrated care is at its best a continuous learning process within the integrated network, resulting in a continuous improvement of the quality of care. To facilitate integrated quality management, a coherent set of methods and models must be taken to plan, organise, operate and evaluate all the many interrelated elements around the care of clients with complex, long-term problems cutting across multiple services, providers and settings. The goals are to enhance the quality of life of the clients, client satisfaction and system efficiency.
This chapter has aimed to find a definition for the quality of integrated care that would make the elements of quality visible for managers, thus also structuring the elements of quality that should be managed. The quality of integrated care should be seen as a result of a quality chain consisting of the elements of care structures, processes and outcomes (Donabedian 1980), and should be evaluated from the perspectives of the client, the professionals, and the provider organisations (Øvretveit 1986, 1992).
This chapter presents some practical examples of quality management approaches that are assumed as fitting to integrated quality management. The models are to be found among those based on total quality management (TQM), such as continuous quality improvement (CQI), as they are multi-actoral and based on structured collaboration. The methods of evaluation of quality of integrated care have been discussed, with a suggested approach of multiple outcome analysis. A special concern has been raised about the lack of both routinely used quality indicators and research initiatives on quality of integrated care. Integrated care particularly needs comparative quality measurement that exploits validated quality and outcome indicators. As the quality indicators for integrated care are lacking, new research initiatives are badly needed to fill the gap.
Careful planning is vital as the initial stage of introducing a scheme for integrated quality management; the way the system is introduced may be even more important than the model chosen. Multi-agency protocols need to be created in collaboration with all relevant stakeholders of the integrated system in question, to define the commonly shared strategy, objectives, methods, measures and protocols on how to ensure good quality care and efficient resource use, and how to organise continuous quality improvement in practical terms. The message is: plan your steps, involve rather than exclude, make sure everyone understands the objectives and means, and make sure that everyone is heard – including the clients and, where relevant, their principal carers.
Dean M (2003). Growing Older in the 21st Century. Swindon: Economic Social and Research Council. Also available at: www.esrc.ac.uk.
Donabedian A (1980). Explorations in Quality Assessment and Monitoring Volume I. Definition of quality. Michigan: Health Administration Press.
Duffy JA, Duffy M, Kilbourne WE (2001). ‘A comparative study of resident, family, and administrator expectations for service quality in nursing homes’. Health Care Management Review, vol 26:3, pp 75–85.
Evers A, Haverinen R, Leichsenring K, Wistow G eds (1997). Developing Quality in Personal Social Services. Concepts, cases and comments. Ashgate, Aldershot: European Centre Vienna.
Huijbers P (2003). De Ontwikkeling van een Zorgprogramma Psychogeriatrie Door de GGZ Noord-Holland Noord te Alkmaar: Een processbescrijving en -analyse (Description and process analysis of two dementia care programmes). Utrecht: NIZW.
Kaplan RS, Norton DP (1996). ‘Translating strategy into action’, in Kaplan RS, Norton DP, Lowes A The Balanced Scorecard. Boston: Harvard Business School Press.
Ministry of Social Affairs and Health/Finnish Association of Local Authorities (2001). Finnish National Framework for High Quality Services for Older People. Handbook 6. Helsinki: MSAH/AFLRA. Also available at: www.stm.fi.
Olsson M (2003). ‘The breakthrough model’ – personal communication.
Øvretveit J (1986). Improving Social Work Records and Practice. Birmingham: BASW.
Øvretveit J (1992). Health Service Quality. Oxford: Blackwell Scientific Press.
Øvretveit J (2001). ‘Quality evaluation and indicator comparison in health care’. International Journal of Health Planning and Management, vol 16, pp 229–41.
Pieper R (2004, forthcoming). ‘Integrated care: concepts and theoretical approaches’, in Vaarama M, Pieper R eds Managing Integrated Care for Older Persons: Perspectives and good practices. Dublin: EHMA/STAKES.
Pollit C (1997). ‘Business and professional approaches to quality improvement: a comparison of their suitability for the personal social services’, in Evers A, Haverinen R, Leichsenring K, Wistow G eds Developing Quality in Personal Social Services. Concepts, cases and comments. Ashgate, Aldershot: European Centre Vienna.
Saskatchewan Commission on Medicare (2001). Caring for Medicare: Sustaining a quality system. Regina: Saskatchewan Health. Also available at: www.health.gov.sk.ca.
Sefton T, Byford S, McDaid D, Hills J, Knapp M (2003). Making the Most of It. Economic evaluation in the social welfare field. Layerthorpe: Joseph Rowntree Foundation.
STAKES (2003). Quality Indicators for Evaluation of Care and Services for Older Persons. Helsinki: STAKES. Available at: www.stakes.fi.
Vaarama M (1995). ‘The EverGreen software for planning services for the elderly’, in Rafferty J, Steyaert J, Colombi D eds Human Services in the Information Age. New York: Haworth Press.
Vaarama M (1997). ‘EverGreen 2000 – a software application for management of the care of elderly people’, in Heumann LF, Miller S eds Managing Care, Risk and Responsibility. Proceedings of the Sixth International Conference on Systems Sciences in Health–Social Services for the Elderly and Disabled. Chicago: SYSTED.
Vaarama M (2004, forthcoming). ‘Integrated planning and evaluation of the care of older persons’, i n Vaarama M, Pieper R eds Managing Integrated Care for Older People: European perspectives and good practices. Dublin: EHMA/STAKES. Also available at: www.stakes.fi.
Vaarama M, Luomahaara J, Peiponen A., Voutilainen P (2001). The Whole Municipality Working Together for Older People. Perspectives on the development of elderly people’s independent living, care and services. Helsinki: STAKES. Also available at: www.stakes.fi.
Vaarama M, Pieper R eds (2004, forthcoming). Managing Integrated Care for Older People: European perspectives and good practices. Dublin: EHMA/STAKES.
Valvanne J (2004, forthcoming). ‘Integrating social and health care in practice – A Finnish example’, in Vaarama M, Pieper R eds, op cit.
www.jcaho.org – website of the Joint Commission on Accreditation of Healthcare Organizations, an organisation working to support performance improvement in health care organisations.
www.mcg.edu/som/fmfacdev/fd_quality.htm – website of the Medical College of Georgia’s family medicine department, which has a section summarising quality assurance.
www.quality.org/TQM-MSI/TQM-glossary.html – website providing a full glossary of total quality management (TQM) terms.
www.sentinel-event.com/focus-pdca_index.htm – website of Medical Risk Management Associates, detailing a range of tools for improving patient safety.
www.stakes.fi – the website of STAKES, the Finnish Research and Development Centre for Welfare and Health, with a range of useful resources and documents. (See individual listings in the references for this section.)