Change search
ReferencesLink to record
Permanent link

Direct link
National Model for Transparent Vertical Prioritisation in Swedish Health Care
Linköping University, Department of Department of Health and Society, Center for Medical Technology Assessment. Linköping University, Faculty of Health Sciences.
Socialstyrelsen, Stockholm.
Linköping University, Department of Medical and Health Sciences, Division of Health Care Analysis. Linköping University, Faculty of Medicine and Health Sciences.
Socialstyrelsen, Stockholm.
Show others and affiliations
2007 (English)Report (Other academic)
Abstract [en]

The proposed national model described in this report has been developed by a working group comprised of staff from the National Board of Health and Welfare, the National Centre for Priority Setting in Health Care, and other organisations involved in vertical prioritisation – including the Östergötland County Council, Stockholm County Council, Västra Götaland, the Health Services Region of Southern Sweden, the Swedish Society of Medicine, the Swedish Society of Nursing, and the Swedish Association of Health Professionals. Throughout the process of designing the model, the Swedish Federation of Occupational Therapists and the Swedish Association of Registered Physiotherapists were regularly informed and given opportunities to review and comment on the proposal. Furthermore, the report was reviewed and discussed at a meeting with invited representatives from the other county councils, the Pharmaceutical Benefits Board, and several professional interest groups. Viewpoints were also obtained at a seminar arranged by PrioNet, a network of individuals interested in prioritisation.

Potentially, the working model described in Chapter 4 could be used in any context where vertical prioritisation takes place, e.g. activities arranged by the state, county councils, municipalities, hospital departments, and professional groups.

This report is designed to be a useful tool for those working on development projects in priority setting. We believe that the contents must be adapted, with the help of relevant examples and some simplifications; to fit the specific needs of different projects or groups. The text must also be adapted to a target group’s knowledge and previous experience in dealing with transparent priority setting. It must be the responsibility of each provider and other affected organisation to adapt the material to the given situation and project. The National Centre for Priority Setting in Health Care, the National Board of Health and Welfare, and others who have participated actively in this effort can be helpful to various target groups in adapting this report.

When and how to engage in practically implementing vertical prioritisation are questions that need to be answered at the local level. Primarily, it is the duty of the local authorities/providers to take responsibility for implementation. Professional organisations also play an important role. Public agencies, universities, and knowledge centres should be sources of support for the local authorities/providers.

The Riksdag’s resolution on prioritisation served as the foundation for developing the model.

Where there are areas of uncertainty in how to translate these guidelines in practice, or where practical implementation might conflict with the principles, we have pointed this out.

Our conclusions and proposals are the following:

  • When facing a choice – regardless of whether it involves allocating new resources for different purposes, or to implement cutbacks – it can be advantageous to rank the possible choices in order of priority. In our model, only the relevant options can be ranked by priority. The consequences of this ranking are not obvious at the outset, but can serve as a basis either to allocate more resources or ration by some means.
  • In vertical prioritisation, it is advantageous to organise the prioritization process starting from a general categorisation of health problems/disease groups. As a rule, these categories cover many organisational units/clinical departments, specialties, or professional groups, thus providing a more multidimensional view of the problem. Furthermore, this allows the process to start from a patient/population perspective, which appears to be more goal-oriented than an organisational/staff perspective.
  • That which is ranked, i.e. one of the choices, we refer to as a prioritization object. We suggest that prioritisation objects consist of different combinations of health conditions and interventions.1 When deciding on the appropriate level of detail, the decision must be based on the context in which prioritisation is carried out. A starting point would be to focus on typical cases, large-volumes services, and controversial care.
  • All forms of vertical prioritisation should be based on the ethical principles that the Riksdag decided should apply in prioritising health services. However, these ethical principles must be made known, clarified, and perhaps complemented before they can be applied to practical priority setting. Furthermore, we believe that the Riksdag’s four so-called priority groups should not be part of the model.
  • The human dignity principle, i.e. that all people should have equal value and equal rights to care irrespective of their personal characteristics and function in society, is the undisputed cornerstone in priority setting. When personal characteristics such as age, gender, lifestyle, or function of a group are expressions of the presence of special needs, so that benefits of the interventions are different, these personal characteristics could be addressed in a priority at the group level. Further discussion is needed regarding the question of how external effects (i.e. the effects of an intervention on families and groups other than the individual directly affected by the intervention) should be valued in priority setting.
  • The concept of need in health care includes both the severity level of the condition and the expected benefits of intervention. As a patient, one needs only those interventions that can be expected to yield benefits. Based on this definition of need, a person does not need an intervention that does not improve health and quality of life, i.e. an intervention with no benefit. In such cases, health services have a responsibility to refer people who seek care for some type of problem, to other appropriate services.
  • The Riksdag’s guidelines regarding the cost-effectiveness principle (applied to individual patients) are too limited to provide guidance for vertical prioritisation at the group level. From the outset, the Government’s bill (Priority Setting in Health Care) highlighted the importance of differentiating a cost-effectiveness principle that applied to choices among various interventions for the individual patient (where the principle can be applied as the Commission of Inquiry proposed) and the aim of health services to achieve high cost-effectiveness in health care generally. Here we also refer to the Riksdag’s directive to the Pharmaceutical Benefits Board. In its decisions on subsidising (prioritising) a drug, the Board should determine, e.g. whether the drug is cost effective from a societal perspective, which requires comparing the patient benefits of the drug to its cost. In such decisions, the cost effectiveness should be considered along with the needs and solidarity principle and the human dignity principle.
  • The proposed working model essentially concurs with the working model used by the National Board of Health and Welfare in developing national guidelines. In describing a national working model, it is not possible to include every aspect that might be considered. Hence, one must start from the model and decide which other relevant aspects should be included. For instance, the International Classification on Functioning, Disability, and Health (ICF) can be used as guidance to describe the severity of health conditions.
  • Due to the wealth of variety in outcome measures for different activities, and the limited experience in working with explicit threshold values, we believe would be premature to recommend standardised categories, e.g. risk levels. However, it is important that those working with prioritization describe their reasoning. Primarily, the categories applied by the Swedish Council on Technology Assessment in Health Care (SBU) to grade the scientific evidence of an intervention’s effects should be used. Local prioritisation projects with limited resources at their disposal should describe (text) their appraisal of the scientific evidence and reference the scientific sources used. The strength of evidence should be expressed in numbers only when supporting a conclusion of a systematic review by SBU, or other literature reviews of good quality.
  • Prioritisation projects having access to health economic evaluation should, until further notice, adhere to the approach used by the National Board of Health and Welfare and present cost-effectiveness on a scale from low to very high cost per life-year gained or cost per quality-adjusted life-year. Economic evidence should be presented according to the principles applied by the National Board of Health and Welfare. In local projects with limited resources, or problems in consistently acquiring information on cost effectiveness, we recommend that the authors at least discuss cost effectiveness in cases where the priority ranking would be decisively affected when costs are weighed in.
  • A 10-level ranking list should be used. The ranking list should be complemented by a “don’t do” list for methods that should not be used at all, or not used routinely, and a research and development (R&D) list for methods where the evidence still insufficient to motivate their use in standard practice. In the absence of an objective quantitative/mathematical method, a qualitative method should be used in the appraisal. Here too, we believe that it is not yet possible to establish standard criteria to determine within which ranking level a prioritisation object should fall.
  • Results should be presented as a ranking list. The parameters used as a basis for prioritisation should also be presented in a uniform manner in ranking lists that are shared with other parties. For pedagogic reasons, details concerning language and format need to be adapted to the respective target groups.
  • Thresholds for what constitutes an acceptable coverage of need (care quality, volume, and percentage of the patient group with access to services) are regional and local issues, and hence are not included in the national model.
Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2007. , 42 p.
National Center for Priority Setting in Health Care, ISSN 1650-8475 ; 2007:1b
Keyword [sv]
Prioritering inom sjukvården, Sverige
National Category
Health Care Service and Management, Health Policy and Services and Health Economy
URN: urn:nbn:se:liu:diva-122805OAI: diva2:873502
Available from: 2015-11-24 Created: 2015-11-24 Last updated: 2016-10-14Bibliographically approved

Open Access in DiVA

fulltext(1071 kB)66 downloads
File information
File name FULLTEXT01.pdfFile size 1071 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Other links

Link to Swedish version and full text

Search in DiVA

By author/editor
Carlsson, PerBroqvist, Mari
By organisation
Center for Medical Technology AssessmentFaculty of Health SciencesDivision of Health Care AnalysisFaculty of Medicine and Health Sciences
Health Care Service and Management, Health Policy and Services and Health Economy

Search outside of DiVA

GoogleGoogle Scholar
Total: 66 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 273 hits
ReferencesLink to record
Permanent link

Direct link