BMJ 1997;315:92-96 (12 July)
Papers
Who should decide?
Qualitative analysis of panel data from public, patients, healthcare
professionals, and insurers on priorities in health care
Karien Stronks, lecturer in public health,a
Anne-Margreet Strijbis, research assistant,a
Johannes F Wendte, senior lecturer in social
medicine,a Louise J Gunning-Schepers,
professor in social medicine a
a Institute of Social Medicine,
Academic Medical Centre, University of Amsterdam, 1105 AZ Amsterdam,
Netherlands
Correspondence to: Dr Stronks
|
|
Abstract |
|
|
Objective: To explore the arguments underlying the
choices of patients, the public, general practitioners, specialists,
and health insurers regarding priorities in health care.
Design: A qualitative analysis of data gathered in a
series of panels. Members were asked to economise on the publicly
funded healthcare budget, exemplified by 10 services.
Results: From a medical point of view, both panels of
healthcare professionals thought most services were necessary. The
general practitioners tried to achieve the budget cuts by limiting
access to services to those most in need of them or those who
cannot afford to pay for them. The specialists emphasised the
possibilities of reducing costs by increasing the efficiency within
services and preventing inappropriate utilisation. The patients
mainly economised by limiting universal access to preventive and
acute services. The "public" panels excluded services that are
relatively inexpensive for individual patients. Moreover, they
emphasised the individual's own responsibility for health behaviour
and the costs of health care, resulting in the choice for
copayments. The health insurers emphasised the importance of
including services that relate to a risk only, as well as feasibility
aspects.
Conclusions: There were substantial differences
in the way the different groups approached the issue of what should
be included in the basic package. Healthcare professionals seem to be
most aware of the importance of maintaining equal access for
everyone in need of health care.
|
Key messages
|
|
|
Introduction |
|
|
As the
costs of health care continue to increase and will do so even more as
our populations age the question of what services a society can
afford to offer to its citizens continues to occasion debate.
Recently readers of the BMJ were confronted with a discussion
of the Swedish report on prioritising health care1 and on Dworkin's thought experiment
on creative rationing.2 Central to this debate of course is
the notion that there are choices to be made. Previous studies have
shown that among the parties involved (public, professionals,
patients), no consensus exists for these choices.3 4 5 This raises the question of whether the
arguments used to defend the different choices are consistent.
To
develop some understanding about the arguments underlying the
opinions of different parties we carried out a study with panels of
five parties that might be concerned in prioritising: the public and
patients, health insurers, and two groups of healthcare
professionals�general practitioners and medical specialists. We
report on their preferences but focus on the criteria developed to
select services to be included in a basic package as well as the
arguments used in specific instances.
|
|
Subjects and
methods |
|
|
The
study took place in the Netherlands. In the Netherlands the (public)
debate on prioritising health care started almost 10 years ago.
Several reports on this issue have been published since then, the
best known of which is the report of the Dunning committee, named
after its chairman Professor Dunning.6 The public debate has been
stimulated by the government�for example, by means of a campaign
aimed at making people aware of the necessity of making choices.
In this
study, six panels (box), consisting of about nine people
each, were asked to play the part of a parliamentary committee
with the decision to select services that would continue to be
funded from a severely cut health budget. The list consisted of 10
different services, selected to represent different aspects of
physical health care (table 1). The decisions to be made were
restricted to those who receive public insurance, which currently
covers about two thirds of the Dutch population.
|
The panels Patients:
People representing
national patient groups The public:
University students
Civil servants not
working on health (None of the members of
these panels suffered from a chronic illness or handicap)
Healthcare
professionals: General practitioners,
in the last stage of their training Specialists in an
academic hospital Health insurers:
|
|
The
panels were given two sources of background information: the
selection criteria as proposed in the Dunning report, Choices in
Health Care: Necessary Care, Effectiveness, Efficiency, Too Expensive
for the Indi- vidual (the criteria are summarised in the
box on the next page)6; and a file on each of the services
under consideration, containing information on its nature, the
number of people affected (now and in the future), costs, and
current restrictions. This information is summarised in table
1. In addition, they received a few papers
stating the views of supporters and opponents of public funding.
|
Dunning criteria Publicly funded
services should meet the following criteria:
|
The
panels were asked to economise nearly one third of the total budget.
They had to decide whether a certain service should be included in
the basic package (that is, provided with public funds) or removed
from the package, totally or partially. More importantly, they had to
give arguments for each choice. They were given five hours to
complete their task. One panel member led the discussion. The
discussions were observed by one of the researchers and a facilitator
structured proceeding as necessary. To check the unanimity of the
outcomes of the discussion the panel members were asked to fill out a
questionnaire afterwards concerning their private opinions. All
discussions were tape recorded and transcribed verbatim.
The
panels were not chosen to be a representative sample of the
population but were selected on purpose. They represented five
distinct parties in health policy. The individual members of the
panels were selected by someone who was familiar with one of the
selected parties. The main selection criteria were: interest in the
subject, willingness to spend a full day on the experiment, and
availability. The "patient" panel was the only one in which the
members were also formal representatives as they were all active
officers of national patient associations. There were two "public"
panels, one of which consisted mostly of university students, who
were thus on average younger than the other panel members. Moreover,
all participants were well educated, which made it possible for them
to assess the information given even though they were not familiar
with it.
|
|
Results |
|
|
All
panels fulfilled their task, although with great difficulty. It was
obvious that most services could not be easily excluded from coverage
in a basic package and that the budget cuts to be achieved were high.
Table 2 summarises the choices made by each
panel.
|
General practitioners
The general practitioners
especially found themselves in a dilemma as they could not really
envision any of the services being reduced without affecting patients
in real need of them. They therefore tried to achieve cost reductions
by limiting access to services, either to those most in need of them
or to those who cannot afford to pay for them. For example, they
discussed restricting travel allowances to low income groups and
decided to pay for in vitro fertilisation only when a woman has no or
only one child. In addition, they decided to restrict the universal
accessibility of home care to nursing care only, but the answers to
the questionnaire they filled in afterwards showed that most of the
panel members felt very uncomfortable about this decision. It is
remarkable that "responsibility of the individual" was not used by
this panel as an argument for exclusion of services. This
underlies the unanimous choice to include treatment of sport
injuries in the benefit package.
Specialists
The medical specialists also thought that
most of the services were necessary. Their criteria for inclusion
were the most explicit: prevention or curing of a disease or caring
for sick people. In their eyes infertility was a disease thus in
vitro fertilisation was included in the basic package. Using that
same argument, however, they went on to exclude oral contraceptives.
A second argument used in that discussion emphasised that most of
the cuts proposed would affect old people. The panel thought
that excluding oral contraceptives was one way to spread the effect
more equitably over age groups. They emphasised possible
economies from increased efficiency within the existing services,
such as home care and prevention of inappropriate use of travel
allowances. Compared with the other panels, the specialist were more
concerned about prevention and hence included breast cancer
screening.
Patients
The patients made a clear distinction
between those suffering from a long term or chronic illness and those
with an acute disease or healthy people. They thought that especially
health services for the former should be publicly funded as
chronically ill people have no other option but to rely on health
care. They therefore firmly rejected cutting costs of long term home
care and travel allowances. On the other hand, they (partly)
excluded preventive services as well as severely cut acute care
(including short term home care after hospital discharge) and health
care needs related to individual behaviour (such as treatment of
sport injuries).
Public
Both panels representing the public looked
very carefully at the argument of financial resources of the
individual and individual responsibility for health. Those services
that were relatively inexpensive or allowed alternative solutions for
the individual were (partly) excluded. These included travel
allowance, oral contraceptives, and in vitro fertilisation. These
choices reflect a high value placed on the principle of individual
responsibility, which was confirmed by the answers given in the
questionnaire. They proposed copayments for health care in the case
of sports injuries and breast cancer screening, a decision
consistent with their earlier arguments.
Health insurers
The panel of health insurers introduced two
criteria specific to their occupational background. In the first
place they separated health risks (for which one can be insured) and
inevitable healthcare needs that will eventually be experienced by
most of us (home care and homes for elderly people). They thought
that the latter should be provided through taxation rather than
through insurance. They also emphasised the feasibility of
implementation of the measures proposed. For that reason they argued
against excluding sports injuries or selecting those in real need for
reimbursement of travel allowances and in favour of the total
exclusion of oral contraceptives.
All panels
Despite these clearly distinct approaches
to the question of priorities, there were also some similarities. All
panels argued that homeopathic medicines should be excluded from
public funding as their effectiveness has not been proved. They all
decided to cut down the costs of the homes for elderly people, the
underlying rationale being the wish to deliver services more
efficiently. In addition, the panels seem to agree on the value of
lifesaving services, as exemplified by lung transplantation. Yet all
groups questioned the cost-benefit ratio of this service, which
reflects the high value attached to efficiency.
|
|
Discussion |
|
|
The
panels were able to reach a consensus, although it was obvious that
many of the concepts used and much of the material offered were alien
to many of the panel members. It is therefore unlikely that an
exercise like this could be easily used to elicit opinions of the
general public. Additional evidence to support this view is provided
by a recent experiment on citizens' juries among the general
public.7 That experiment showed that if the
general public is to be involved in decisions on prioritising in
health care they need much more time and information than the
five hours and the written information that the participants in
our study were given.
Given
the fact that the panel members were relatively young and highly
educated, they were far from random samples. Therefore the panels
should not be considered as representing the whole population of
patients, general practitioners, etc. It could be questioned whether
this threatens the validity of our results. In general demographic
characteristics do not seem to be systematically related to the
decisions people make in prioritising processes.8 An exception should probably be made for age as
a recent study showed that people seem to favour services for their
own age group.5 If this is true, the panels probably would have
been more reluctant with respect to cutting costs for home care and
homes for the elderly if more elderly people had been included. On
the other hand, a specific check on the age distribution of the
proposed cuts was discussed explicitly by at least one of the panels.
Moreover, our aim was to obtain more insight into the arguments
underlying the choices of different actors in the decision making
process, rather than into the outcome of the prioritisation
process. We do not consider it likely that the arguments strongly
differ by age or other characteristics.
In view
of the limited share of resources that can be devoted to health care,
all panels chose to exclude ineffective services from public funding
and tried to deliver services more efficiently. Hence the idea of
effectiveness and efficiency as necessary conditions for the general
accessibility of our future healthcare system seems widely shared.
This was confirmed by the data obtained by the questionnaire. In
addition, all groups seemed to agree on the value of lifesaving
technologies. The discrepancy with the results of other studies
suggesting that the public and professionals disagree in this
respect3 9 might be explained by the fact that
we chose lung transplantation as an example of high technology, a
service which mostly benefits young people. Moreover, the costs of
this service represented only a small proportion of the total
budget.
Observed differences
Besides these similarities we
observed substantial differences in the arguments of the groups
considered. Patients and health insurers in particular explicitly
introduced self interest as a basis for prioritising. The patients,
all suffering from a chronic illness, considered care services to be
the most important, which corresponds with the results of another
study regarding primary healthcare services.4 The choices of the health insurers
were guided mainly by their concern for the practical feasibility
of the chosen strategy. We found little evidence to suggest
that the public or healthcare professionals were guided by personal
needs or self interest, which is consistent, as far as the public
is concerned, with the results of a previous study.10
Partly
because of differences in the approach to the population's best
interest, but probably also because of differences in first hand
experience with the people using the different services discussed,
the main difference between the panels seems to be the extent to
which they took the principle of equal access into consideration.
This finding seems to be in accordance with that of a previous
studies.3 11 Members of the public panel in particular
frequently emphasised the importance of individual responsibility.
In accordance with this principle, copayments were proposed
without paying much attention to whether this might affect the
accessibility of health care for low income groups. Although
the outcome here may have been biased by their relatively high
level of educational and income, the results of a survey among
a random sample of the public3 also showed that priorities set by
the general public might be contrary to the principle of equity and
equal access. In addition, while focusing on a conception of health
care in terms of care for the chronically ill, the strategy of the
patients might threaten the accessibility for those with an acute
illness or healthy people. In contrast, the healthcare professionals,
and in particular the general practitioners, explicitly took into
account the consequences of their decision for the equal
accessibility of services. In addition, the specialists discussed the
distribution of the proposed economies across age groups.
Finally, the healthcare professionals were the
most pessimistic about the possibilities of cutting down public
expense on health care. The strategy of not offering those services
under public funding was hardly an acceptable option to them, whereas
the possibilities for other strategies, such as restricting
access to specific groups, was limited by the high value attached
to the principle of equal access. A cynic might argue that
doctors would be expected to be most reluctant to cut healthcare
costs out of self interest, but the arguments used and the tone
of the discussion suggest otherwise. The difficulties
encountered by the healthcare professionals in the decision making
seem to reflect a more realistic picture of current practice in
health care, as well as a greater awareness of the importance of
maintaining equal access.
From
our results it is not clear that including all the different actors
in the decision making process of prioritisation of health services
will lead to more equitable or broadly supported outcomes or to
better health for the population. It is quite remarkable that it is
the medical profession that seemed most concerned about the common
good and the distribution of services. The panels representing the
public, patients, and insurers had much more trouble weighting the
interests of all those concerned. As a result, their decisions might
threaten the universal accessibility of core services. As this
principle is highly valued in western societies this may be
unacceptable for at least professionals, as our results show, but
also for policy makers. The arguments put forward by the different
groups, however, are well worth considering because they clearly
reflect the different perspectives of those affected by
prioritisation. We think that they all deserve to be carefully
considered by the (democratically elected) bodies that decide on
resource allocation in most of our healthcare systems.
|
|
Acknowledgements |
Funding: This project was partly funded by a
grant from the Dutch ministry of public health.
Conflict of interest: None.
|
|
References |
|
|
- McKee M, Figueras J. Setting
priorities: can Britain learn from Sweden? BMJ 1996;312:691-4.
[Free Full Text]
- Smith R. Being creative about
rationing. BMJ 1996;312:391-2.
[Free Full Text]
- Bowling A, Jacobson B, Southgate
L. Health service priorities. Explorations in consultation of the public and
health professionals on priority setting in an inner London health district.
Soc Sci Med 1993;37:851-7.
- Hopton JL, Dlugolecka M. Patients'
perceptions of need for primary health care services: useful for priority
setting? BMJ 1995;310:1237-40.
[Abstract/Free Full Text]
- Bowling A. Health care rationing:
the public's debate. BMJ 1996;312:670-4.
[Abstract/Free Full Text]
- Government Committee on Choices in
Health Care. Choices in health care. Rijswijk: Ministry of Welfare,
Public Health and Cultural Affairs, 1992.
- Lenaghan J, New B, Mitchell E.
Setting priorities: is there a role for citizens' juries? BMJ
1996;312:1591-3.
- Fowler F, Berwick DM, Roman A,
Massagli MP. Measuring public priorities for insurable health care. Med
Care 1994;32:625-39.
- Tymstra T, Andela M. Opinions of
Dutch physicians, nurses, and citizens on health care policy, rationing and
technology. JAMA 1993;270:2995-9. [Medline]
- Dicker A, Armstrong D. Patients'
views of priority setting in health care: an interview survey in one practice.
BMJ 1995;311:1137-9.
[Abstract/Free Full Text]
- Lomas J. Reticent rationers:
consumer input to health care priorities. In: Gunning-Schepers LJ, Kronjee GJ,
Spasoff RA, eds. Fundamental questions about the future of health care.
Gravenhage: Wetenschappelijke Raad voor het Regeringsbeleid, 1996.
(Accepted 10 April 1997)