What is a priority?

Abstract

What does it mean to say that something is a 'priority'? Priority setting is used to balance competing claims for resources, but the nature of the exercise is ambiguous. The priorities which are claimed might be for time, resources, process, rights or service. The setting of priorities might refer to importance, relative value, precedence, special status or lexical ordering. And there are different ways of ranking priorities including simple ordering, optimization, triage and satisficing. There is a fundamental distinction between preference rankings and precedence rankings, which can lead to strongly different conclusions from the same initial information. Because there is no definitive understanding of what a priority is, there can be no authoritative formula for deciding between competing claims.

Full text

OpenAIR@RGU
The Open Access Institutional Repository
at Robert Gordon University
http://openair.rgu.ac.uk
This is an author produced version of a paper published in
Journal of Health Services Research & Policy (ISSN 1355-8196, eISSN
1758-1060)
This version may not include final proof corrections and does not include
published layout or pagination.
Citation Details
Citation for the version of the work held in ‘OpenAIR@RGU’:
SPICKER, P., 2009. What is a priority? Available from
OpenAIR@RGU. [online]. Available from: http://openair.rgu.ac.uk
Citation for the publisher’s version:
SPICKER, P., 2009. What is a priority? Journal of Health Services
Research & Policy, 14 (2), pp. 112 -116.
Copyright
Items in ‘OpenAIR@RGU’, Robert Gordon University Open Access Institutional Repository,
are protected by copyright and intellectual property law. If you believe that any material
held in ‘OpenAIR@RGU’ infringes copyright, please contact openair-help@rgu.ac.uk with
details. The item will be removed from the repository while the claim is investigated.

What is a priority?
Paul Spicker
Journal of Health Services Research and Policy 2009 14(2) 112-6
Abstract. What does it mean to say that something is a “priority”? Priority setting is used
to balance competing claims for resources, but the nature of the exercise is ambiguous.
The priorities which are claimed might be for time, resources, process, rights or service.
The setting of priorities might refer to importance, relative value, precedence, special
status, or lexical ordering. And there are different ways of ranking priorities within
different understandings of the term, including simple ordering, optimisation, triage, and
satisficing. There is a fundamental distinction to be made between preference rankings
and precedence rankings, which can lead to strongly different conclusions from the same
information base. Because there is no definitive understanding of priority, there can be no
authoritative formula for deciding between competing claims.
The idea of prioritisation is associated with the exercise of judgement between competing
claims. Part of the literature on priority setting in health care is concerned with the political
process of negotiating for resource allocation (e.g 1, 2), and there is a specialised literature
concerned with the methods used to distinguish between priorities. In both, “priorities” are
mainly identified by establishing the preferences and views of stakeholders (practitioners,
managers or the public). Ham, for example, identifies “priority setting” partly with the
kind of initiative established in Oregon, the Netherlands and the UK (3) - which attempt to
establish criteria for funding - as well as schemes in New Zealand and Sweden, which are
attempts to choose between conflicting claims. Some studies focus on establishing a
framework for eliciting and drawing together competing views, such as programme
budgeting and marginal analysis (PBMA)(4), Multi-Criteria Decision Analysis
(MCDA)(5) and even qualitative analysis based on grounded theory (6). Studies seem to
assume that the main issue in setting priorities is to establish what stakeholders think is
important, or some kind of criteria for evaluation: once values and criteria have been
established, the setting of priorities in some sense follows (e.g. 7) Balancing competing
claims is consequently a technical exercise: examples include discrete choice modelling
(8), conjoint analysis (9), or various forms of decision analysis (10,11).
The purpose of this paper is not to review this kind of method, or to examine the political
process, but to ask a simpler, more basic question. What does it mean to say that
something is a “priority”? There are several different understandings of the term, and
unless we can work out what priority setting is supposed to do, a system which sets out to
establish priorities is unlikely to reflect the issues and concerns it is intended to resolve.
Priority of what?
A useful initial illustration of priority ranking might be one of the general statements
1

reviewed by Ham (3). This was the order of priorities defined by a parliamentary
commission for clinical treatment in Sweden:
“1a. Treatment of life-threatening acute diseases. Treatment of diseases which if
left untreated will lead to permanent disability or death.
1b. Treatment of severe chronic diseases. Palliative terminal care. Care of persons
with reduced autonomy.
2. Individualised prevention during contacts with medical services.
Habilitation/rehabilitation [as required by Swedish law]....
3. Treatment of less severe and chronic diseases.
4. Borderline cases.
5. Care for reasons other than disease or injury.”
The list is partly about the allocation of resources, but there is no implication that the
allocation of resources will be proportionate to priority. Priority is, rather, a claim to
receive resources before lesser priorities do. In the clinical setting, it is a claim to be dealt
with first - which does not mean that lesser priorities will not be dealt with, but that they
may be dealt with more slowly, or may even be set aside to meet the demands of higher
priorities.
Claims for priority relate to a wide range of factors. Priority for resources might refer to
financial allocations, staff time or expertise, agency capacity and physical resources (like
bedspace). Priority for time can be considered in three ways: in terms of relative urgency,
total time allocation (which is time as a resource), or when things are done (the Swedish
review gives the example of pre-operative medication, which has to be done at a specific
point in time before other work). There may be claims on process: for example, that a
person is given an opportunity (such as a right to choose between practitioners or locations
for treatment) or, in the name of empowerment, the space to express views. There may be
priority of rights - for example, the right of one claimant to make a choice before another,
or the priority which makes someone in worse circumstances have to wait while someone
with a lesser need is still in the bed. Priority is often claimed for people with greater needs,
but in a rights-based service, greater needs are not necessarily more important than other
needs - they are simply bigger. And, in terms of the operation of rationing processes, there
may be priorities of service, like professional standards or functional specialism. Any new
allocation has to meet the criteria which guide the service, ensuring that the service is
operating consistently with its objectives.
What priority setting is doing is ambiguous from the outset. The purpose may be to
manage and judge between competing claims, but unless there is a specific context, clear
identification of the issues to be resolved and an understanding of what the competing
claims are, there is little prospect of devising a priority ranking that makes any kind of
sense. The key objection to the kind of exercise referred to at the beginning of this paper
is, however, even more fundamental. The literature generally assumes that if we have to do
is to ask professionals what is important, we can feed the results into a computer, and that a
priority ranking will follow. If we do not know what a "priority" means, we cannot.
Five types of priority
2

Priority as importance. The most basic meaning of a “priority” is as something which is
more important than something else. If we look at the way that the term “priority” is used
in practice, the “something else” is often vague.
“Tackling health inequalities is a top priority for this Government” (12)
“Tackling obesity is a government wide priority.” (13)
“It is a top priority for my department to reduce the risk of infection”(14)
“Tackling heart disease remains a top priority.”(15)
“Chronic disease management is currently a top priority for the NHS”(16)
“Being open and fair must become a top priority in healthcare.”(17)
Saying that something is a “top priority” does not seem to say much more than that the
topic matters, and something should be done. It is tempting to dismiss this usage as trivial -
but if it means nothing, why do politicians spend so much time saying it?
By contrast with “priority” (or even “top priority”), describing something as “not a priority”
does have a direct implication for practice. The Mayor of London’s Office complains:
“Among the services threatened [with cuts] after being determined ‘not a priority’
are risk-reduction interventions for children and young people involved in or at risk
of sexual exploitation; reducing the involvement of young people in violent crime
(particularly the prevention of involvement in gang culture, knife and gun crime);
support for women and children affected by domestic violence through the
provision of independent advocacy; and support for increased play and physical
activity for children.” (18)
If all of these things were “priorities”, along with everything else that the local authorities
do, there would be a very long list indeed. The distinction seems to lie between those
activities which government is prepared to undertake, and those which it is not.
However, this is not the only way to treat priorities. It should be possible to accept, for
example, that antenatal care or elective surgery are not “priorities”, in that other services
currently have stronger claims for resources, while at the same time seeking to maintain or
improve the services to some degree. In an environment where “not a priority” is often a
euphemism for “subject to closure”, this presentation has become unusual, but there are
occasional examples. The UK Health and Safety Commission states that:
“While HSE will continue to promote sensible advice to employers ... work-related
road safety is not a priority for HSC/E.”(19)
Saying that something is not a priority is not - or should not be - the same as saying it will
not be done; it is saying it will be done less than other things. A fuller understanding needs
to consider priorities relative to other things - to lesser priorities, or to non-priorities.
Relative value. Saying that something is important says only that it will get more weight
than something else; this could be simply a decision to spend more money on one thing
than another. The principal alternative to priority as “importance”, exemplified in the kind
of optimisation model mentioned at the beginning, is that a relative value or weight is
attached to each factor, and resources are allocated proportionately to that value. Even if
there is a clear preference for certain priorities over others, it is not self-evident that the
highest priority should take precedence over lower priorities in every case. Sen and
Paterson-Brown argue, in the context of obstetric care, that
3

“It is important to keep resolving the minor cases and to anticipate and sort out the
intermediate problems so that the serious emergencies are less likely to occur, and
so that when they do occur staff are prepared and able to deal with them.”(20)
The factors which are being given priority may be divisible, like resources; they may be
capable of being balanced, like waiting time. Priorities are relative, not absolute; the
highest priorities do not drive out lower priorities. Typically, non-urgent surgery is subject
to more limitations than urgent surgery, but both are still done - which means that there will
be occasions where the less urgent surgery is done in preference. The preferences that
people express may not be for absolute priority, but for balance.
A priority with greater relative value can be seen as something which has higher
utility, or is chosen at greater cost than, a lesser priority. The methodology of economic
evaluation commonly seeks to present priorities in terms of resource allocation, balancing
competing choices in appropriate combinations. Economic theory generally works on the
proposition that collective choices are simply the aggregate of individual choices, and any
allocation is the outcome of the sum of such choices. Any collective result will
consequently combine different outcomes, rather than giving precedence to any single
factor. The establishment of priorities is equivalent in practice to the allocation of a
household budget; choices are optimised in terms of the relative utility of different options,
and the expression of priorities is understood in terms of their relative allocations.
There is a subtle but important difference between the economic, choice-based
model of preference and the pattern of preference generally adopted in priority-setting
exercises. It is always possible to arrange them mathematically so that they yield the same
results overall. However, the economic model is bottom-up; allocations result from
aggregate choices, and everyone gets the balance they choose within the resources they are
allocated. Weightings based on collective preferences are top-down; the budget allocation
reflects the average rather than the aggregate. Some budget choices are mutually exclusive,
or at least made between alternatives - for example, the choice of a specific location for a
service, a decision to develop a speciality or to generalize, or choices of treatment for the
same conditions. Where allocative decisions are being made on the averaged preferences,
the resulting compromises may satisfy no-one.
Precedence. The next alternative interpretation is that the priority has precedence over the
non-priority, so that although both are considered, the higher priority has to be dealt with
before the lower priority can even be considered. This has the implication that when
resources are scarce, priority will lead to allocation to a higher priority to the exclusion of
lower priorities. In the allocation of housing, for example, each house goes to the next
person in priority order.(21) If there are three houses, three households will be rehoused; if
there are four, four people will be. The distinction between third and fourth place - and,
indeed, for all rankings - is specifically meaningful, and for the households in question it
may be crucial. It also implies that the effect of treating homelessness as a priority over
living in bad conditions is to imply in every case that people in bad conditions will not be
housed unless there is no homeless person with priority over them. The list of health
priorities from Sweden is arguably a priority list of the same kind. The implication of a rule
based on precedence is that it is questionable whether any resources at all should be
available for “borderline cases” when people with higher priorities need treatment.
This is probably not, however, a fair reflection of the way priorities usually work in
health care. In the UK context, if someone has cancer or heart disease, and those
4

conditions are deemed to have the highest priorities, people with those conditions still have
to wait longer because medical resources are being used for someone who needs a hip
replacement, and few people would want to create a situation where lesser conditions could
not be treated. There is a further principle at work here: hip replacements or borderline
cases are protected because there is a rule about medical care, which has its own priority,
distinct from the priority given to other diseases.
Priority as special status. Priority setting may mean that where principles come into
conflict, protecting the priority area will determine the outcomes. Priorities, by this test,
should include issues which are reserved, guaranteed, protected or ring-fenced. An
example of this form of prioritisation is the practice of “mainstreaming”. (22)
Mainstreaming means that a designated priority issue always has to be taken explicitly into
account, even if the decision is supposed to be about something else. In the European
Union, the impact of policies on gender and the environment usually have to be considered
in this way. The effectiveness of this practice can be questioned - too often, mainstreamed
topics can be considered superficially or ritualistically - but it is difficult to deny that it is
giving them a special status and priority.
Although “mainstreaming” may seem rather specialised, this approach to priorities
might be more prevalent than it first appears. Policy-making often centres on principles
which are ill-defined and inexplicit; they come into play only when they are violated. By
way of illustration, the development of strategic planning and partnership working has
given considerable prominence to new agendas, including for example community safety,
health improvement and sustainability. These are the explicit priorities, but being explicit
does not mean that they are the most important. Some higher priorities tend not to be
mentioned directly: they include issues like service delivery, cost-effectiveness or
professional standards. These issues will not have been forgotten, but it does mean that
discussion and policy development seems to proceed without them being mentioned at all -
until it becomes apparent that there is an issue, and the true priorities re-surface.
Lexical ordering. Lastly, a priority might imply simply that an issue has to be tackled first,
before anything else can be. John Rawls used the term “lexical” for issues which need to
be considered before other issues can be.(23) If an issue is lexically prior, it has to be dealt
with before other issues, and this will be true even if it involves fewer resources or is
ostensibly less important than others. Hospital care may carry more weight politically than
primary care, but unless conditions are diagnosed and referrals are made, the hospital is not
going to serve its purpose; primary care has to come first.
Lexical ordering does not imply that the first priorities will receive more resources
than others; the weight and proportions are not necessarily the same as the order in which
things are done. That should also mean, in principle, that lexical ordering could take place
simultaneously with other patterns of priority setting.
Preference and precedence rankings
Although there are five discrete understandings of priority here, two (priority as
importance, and priority as special status) have limited direct impact, and lexical ordering
commonly occurs in combination with others. The two core definitions, relative value and
5

precedence, are the ones that matter most. There is a key distinction to make between
preference and precedence ranking.
Priority setting is generally used as a guide to the use of limited resources. If resources are
held constant, it is usually possible to model different understandings of the priorities to
produce similar allocative results. As resources change, however, the implication of
different priority rankings or scales becomes very different. In a preference ranking, if a
large percentage of resources is devoted to the top priorities and a lesser percentage goes to
second-order priorities, the usual implication of an increase or decrease in resources will be
that the change is distributed in the same proportions. So, if there is an initial allocation
between two priorities of 70% to 30%, and resources increase by 10%, in a preference
ranking the new allocation would be 77:33. This general principle applies to priorities
distributed according to importance, choice or relative value. In a precedence ranking,
where one priority is considered more important than others, the higher priority is dealt
with first; when resources are limited, the higher priority will be protected, and resources
will be denied to the lower priority; when resources increase, the extra resources should in
principle be distributed to the lower priority (because it was denied the resources in the first
instance). The preference ranking distributed resources 77:33; in a precedence ranking the
new distribution should be 70:40. Conversely, a reduction of 10% would mean a
difference between 63:27 in a preference ranking, and 70:20 in a precedence ranking.
In real life, any decision would almost certainly be tempered by other considerations, like
value for money, the impact on services and an assessment of how long the resources
would last, but the implications are clear: these are fundamentally different outcomes
arrived at on the basis of the same initial information. The difference is made by the
definition of priority that is adopted.
Methods of prioritisation
Even within the different understandings of priority, there are different ways of putting
priorities in order. Some options are implicit in the preceding discussion. In a precedence
ranking, the most obvious thing to do is to take the highest priority first: higher priorities
exclude lower ones until sufficient resources have been allocated to meet the higher
priority. In the context of a preference ranking, the simplest approach is to allocate
resources according to their relative value.
However, there are different approaches to optimisation. Linear programming offers
techniques for short-term optimisation of conflicting priorities, while other forms of
decision analysis offer the means to include longer-term or speculative considerations. For
the most part this kind of decision making process tends to assume a preference ranking
based on relative weights, but there have been circumstances where this type of weight is
used to produce a points scheme, and the points scheme is then used to make a precedence
ranking.
Another option is triage. Triage is often a pragmatic strategy, concerned with sifting and
sorting presenting problems in order to direct patients to appropriate tracks of care.(24, 25)
6

The term is also used, however, for the process of prioritising people’s circumstances in
order to achieve the most effective outcomes with limited resources:
The purpose of the triage process in a major emergency is to ensure that limited
time and other resources available are used to care for those who will most benefit,
rather than for those with minor injuries or those who have little chance of
survival.(26)
Triage, in other words, balances considerations of need with other priority considerations -
such as ability to benefit, relative cost or ease of access. Priority rankings can consequently
be revised to maximise effectiveness.
Fourth, there is satisficing. Policies are commonly guided not by priorities but by a
combination of pragmatism and an acceptance of the limitations of data sources. If health
improvement is a priority, there would be little reason to focus on specific areas, localities
or schools; most disadvantaged people do not live in disadvantaged areas, and most of the
people in disadvantaged areas are not themselves subject to the conditions being focused on
in policy. Area-based policies and service location are typically formed in the light of
indicators, consideration of the practicalities of locating and delivering services, and a
judgment about proportionate effectiveness to effort.
The implication of these arguments is that any attempt to determine priorities simply by
ranking priorities, allocating weights or ordering preferences is unlikely to be satisfactory.
Formal methods of reconciling views typically depend on a series of assumptions about
relative weights and procedures which may bear little resemblance to the factors
influencing priority setting in practice. There is no definitive understanding of priority, and
consequently there can be no authoritative formula for deciding between competing claims.
Notes
1. Gibson J, Martin D, Singer P. Priority setting in hospitals. Social Science and Medicine
2005; 61(11): 2355-2362
2. Vuorenkoski L, Toiviainen H, Hemminki E. Decision making for priority setting in
medicines. Health Policy 2008; 86(1):1-9
3. Ham C, Priority setting in health care: learning from international experience. Health
Policy. 1997; 42: 49–66
4. Mitton C, Donaldson C. Setting priorities and allocating resources in health regions.
Health Policy 2003; 81(3): 335-348
5. Mabin V, Menzies M, King G, Joyce K. Public sector priority setting using decision
support tools. Australian Journal of Public Administration 2001; 60(2): 44-59
6. Singer P, Martin D, Giacomini M, Purdy L. Priority setting for new technologies in
medicine. British Medical Journal 2000; (November): 1316-8
7. Olsen J, Donaldson C. Helicopters, hearts and hips: using willingness to pay to set
priorities for public sector health care programmes. Social Science and Medicine 1998;
46(1): 1-12
8. Farrar S, Ryan M, Ross D, Ludbrook A. Using discrete choice modelling in priority
setting. Social Science and Medicine 2000; 50(1): 63-75
9. Oudhoff J, Timmermans G, Knol D, Bijnen A, van der Wal G. Social science and
medicine 2007; 64:1863-1875
7

10. Dror D, Koren R, Ost A, Binnendijk E, Vellakkal S, Danis M. Health insurance benefit
packages prioritized by low-income clients in India. Social Science and Medicine 2007;
64(4): 884-896
11. Peacock S, Richardson J, Carter R, Edwards D. Priority setting in health care using
multi-attribute utility theory and programme budgeting and marginal analysis. Social
Science and Medicine 2007; 64(4): 897-910.
12. Department of Health. Health Inequalities,
http://www.dh.gov.uk/en/Publichealth/Healthinequalities/index.htm, accessed 25.2.08
13. Department of Health. Obesity,
www.dh.gov.uk/en/Publichealth/Healthimprovement/Obesity/index.htm, accessed 25.2.08
14. Department of Health. Departmental Report 2005, Cm 6254, 2005,
http://www.dh.gov.uk/en/PublicationsAndStatistics/Publications/AnnualReports/DH_4113
725 accessed 25.2.08
15. Department of Health. The National Service Framework for coronary heart disease:
winning the war on heart disease, 2004.
http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGui
dance/DH_4077154 accessed 25.2.08
16. Pharmacy Flyer 2004, Community pharmacy network gives UK a head start on
chronic disease management, 2004; 15:1.
http://www.natpact.nhs.uk/uploads/Pharmacy%20flyer%20summer%202004.pdf accessed
25.2.08
17. National Patient Safety Agency. Seven steps to patient safety. 2004.
www.npsa.nhs.uk/EasySiteWeb/GatewayLink.aspx?alId=5635 accessed 25.2.08
18. Mayor of London. Grants for London’s children’s services under threat. 25 September
th
2006, obtained at http://www.egovmonitor.com/node/7758 accessed 25.02.08
19. Health and Safety Executive. Work related road safety.
http://www.hse.gov.uk/roadsafety/ accessed 25.2.08.
20. Sen R, Paterson-Brown S. Prioritisation on the delivery suite, Current Obstetrics and
Gynaecology. 2005; 15: 228-236.
| 21. Mullins D, Pawson H. The land that time forgot. Policy and Politics 2005; 205-330.
22. Spicker P. Policy analysis for practice, Bristol: Policy Press 2006.
23. Rawls J. A theory of justice. Oxford: Oxford University Press 1972.
24. Roy P, Lannehoa Y. Le triage aux urgences. Réanimation 2002;(11): 480-5
25. N Sands N. Mental health triage. Journal of Psychiatric and Mental Health Nursing.
2007; 14:243-249
26. Scottish Executive Health Department, NHS Scotland - Emergencies,
http://www.sehd.scot.nhs.uk/EmergencyPlanning/Documents/part_3.htm accessed 25.02.08
8