Opinions of Swedish citizens, health‐care politicians, administrators and doctors on rationing and health‐care financing

Abstract

Objective  To compare the views of citizens and health‐care decision‐makers on health‐care financing, the limits of public health‐care, and resource allocation.

Design  A postal survey based on a randomized sample of adults taken by the national registration and stratified samples of health‐care politicians, administrators, and doctors in five Swedish counties.

Participants  A total number of 1194 citizens (response rate 60%) and 427 decision‐makers (response rate 69%).

Results  The general public have high expectations of public health‐care, expectations that do not fit with the decision‐makers' views on what should be offered. To overcome the discrepancy between demand and resources, physicians prefer increased patient fees and complementary private insurance schemes to a higher degree than do the other respondents. Physicians take a more favourable view of letting politicians on a national level exert a greater influence on resource allocation within public health‐care. A majority of physicians want politicians to assume a greater responsibility for the exclusion of certain therapies or diagnoses. Most politicians, on the other hand, prefer physicians to make more rigorous decisions as to which medical indications should entitle a person to public health‐care.

Conclusions  The gap between public expectations and health‐care resources makes it more important to be clear about who should be accountable for resource‐allocation decisions in public health‐care. Significant differences between physicians' and politicians' opinions on financing and responsibility for prioritization make the question of accountability even more important.

In Sweden, as in many other Western countries, priority‐setting in health‐care is in conflict with citizens' expectations and patients' demands. ¹ , 2, ^–3 From a political equity‐based perspective, treatments which are of low medical value and/or are associated with high costs are less desirable than more cost‐effective ones, and resources should be used where the greatest benefits are to be found. ⁴ Even if different patient groups are rarely placed in opposition to each other, the resource‐allocation system reflects these choices, often in the shape of queues. Treatments directed at trivial needs, or needs not necessarily regarded as medical needs, are also questioned in the health‐care debate. At the same time, patient demands are increasing as a consequence of the expectations born of major developments in medical technology.

Few health authorities seem to know what the expectations and views of the public are about health‐care priorities. ⁵ Not only are such views of great importance in themselves, they should also be used as a balance against the opinions of county‐council health‐care politicians and service providers. ⁶ Usually, these groups are examined in isolation, but some comparative studies have been conducted both in Sweden and in other countries. ⁷ , 8, 9, ^–10 In most cases, these comparative studies amount to rankings of health‐care services or patient needs, or evaluations of different prioritization criteria. ¹⁰ , 11, ^–12 Less interest has been paid to financing and decision‐process issues.

Politicians are also reluctant to address the issue of additional resources, but pin great hopes on the introduction of evidence‐based medicine, guidelines, and more cost‐effective production. ¹³ Their confidence in the doctors' ability to prioritize in a fair and rational way is strong, and the widespread strategy has been to `ration by muddling through elegantly'. ¹⁴ However, in April 1997 a decision on guidelines for priority‐setting was taken by the Swedish Parliament. In order to maintain a national monitoring mechanism for priority‐setting, Parliament also decided to set up a new national body for priorities in health‐care, the National Priority Commission. ⁶ The task assigned to it has been one of stimulating discussion and gathering information. Within that framework, a number of activities are being taken forward and it was at the Commission's request that the present study on opinions among decision‐makers and citizens was carried out. The main aim of the study was to investigate differences in attitudes between citizens, physicians, health‐care politicians, and administrators. A series of questions was selected to elucidate (a) the public's expectations of public health services (as compared with those of the experts) and (b) the views of the different expert groups on influence and responsibility in priority‐setting.

Method

Design of questionnaire

The design of the questionnaire and implementation of the study were carried out in accordance with the total design method (TDM), a method that generally produces high response rates. Following the method means strict compliance to certain guidelines for question construction, layout, enveloping, follow‐ups, etc. ¹⁵ Reminders were used and one postcard was sent out after 2 weeks and a new questionnaire after another week. Most of the questions were given in the form of statements that respondents were asked to respond to. The questions were classed in four categories: distribution of health‐care resources, drugs, decision‐making and ethical aspects on priorities. The complete results from all questions have been reported elsewhere. ¹⁶

Study areas

The five study counties have populations between 250 000 and 400 000, and they were strategically selected based on geographic and demographic properties as well as their varying experiences of structured prioritization strategies. Three of the counties were included in a previous study. ⁹

Study groups

In the pilot study a test questionnaire was sent to 300 individuals randomly selected from four study populations. In the main study citizens and physicians were randomly selected from relevant registries, while the group of politicians and administrators consisted of the most senior decision‐makers at regional and district levels in the five counties. The physicians were equally split into heads of departments and general practitioners. A total number of 2604 questionnaires were delivered, a sample large enough to permit valid comparisons between several subgroups.

The response rate was 60% from the general public, 80% from the politicians, 60% from the physicians and 73% from the administrators. Distribution between study groups is given in Table 1. The non‐respondents do not cause any trouble as the groups are well profiled, the comparisons are relatively few and the numbers large enough to permit some longitudinal reliability tests.

Table 1.

  Study groups (number and percentage)

Statistical methods

Descriptive statistics was used to characterize different strata in the population, and a test of the differences was carried out using non‐parametric methods (chi‐square). To avoid age‐related distortions, comparisons between study groups were only conducted in the age span 30–65 years. The primary purpose of the study was to compare different strata in the population, and less interest was paid to aggregate numbers.

Results

The selected results from the survey are presented in two subsections. The first covers preferences related to unlimited public health‐care and private funding, and the second presents views on the resource‐allocation process. In the first subsection, the public's opinions are compared with those of physicians and health‐care administrators and politicians, and the second subsection compares the three decision‐maker groups. The reason for not collecting the citizens' views on the resource‐allocation issue was the judgement that informed answers imply practical knowledge of the decision‐making process.

Access to public health‐care and private funding

There was a marked difference between citizens and decision‐makers when it comes to their views on whether the public health services should always offer the best possible care, irrespectively of cost (P < 0.001, d.f. 3). The largest difference was between citizens and physicians: 59% of citizens and only 12% of physicians agreed fully with the statement (Table 2). The study results could be compared with an earlier study on 1574 primary‐care patients in Sweden, where 78% agreed fully with the same statement. ¹⁷

Table 2.

  Politicians', administrators', physicians' and public opinions on the public health‐care commitment (%)

A second statement in the recent study asserted that every individual has a right to have his or her health‐care needs met, even if troubles are trivial (Table 2). The views of the decision‐makers and the public differed (P < 0.001, d.f. 3), and especially administrators and physicians were more apt to oppose the statement. Forty percentage of citizens fully agreed, but for example only 5% of administrators. In the above‐mentioned patient survey, 71% of the patients fully agreed. ¹⁷

Do physicians, administrators and politicians hold other views than citizens as regards funding issues as well? Two questions sought to analyse this (Table 2). Where more money would be needed to maintain present standards in public health‐care, decision‐makers and the public held different views as to which type of financing should be used. Health‐care politicians in particular favoured increased taxes, whilst physicians were more reluctant. They, on the other hand, were more favourably disposed than the other respondents to complementary private insurance and increased patient fees.

The same pattern emerged when the question concerned the problem of growing numbers of elderly people needing medical treatment and care against a decreasing number of taxpayers (Table 2). Decision‐makers and citizens had different views on the financing issue. Politicians tended to be more in favour of increased taxes, and physicians preferred complementary private insurance to a higher degree than the other respondents did.

Resource allocation

The decision‐makers were asked for their opinion as to who has the greatest influence on resource allocation within public health‐care today (Table 3). The answers differed between politicians, administrators and physicians (P < 0.001, d.f. 12); but it should be noted that as a result of the relatively high number of response alternatives, the separate percentages only correspond to a limited number of respondents. In all three groups, the regional health‐care politicians were seen as the most influential actors. When choosing the second, or third, most influential group, views differed.

Table 3.

  Politicians', administrators' and physicians' opinions on prioritization in public health‐care (%)

When the question concerned who should have the greatest influence on resource allocation within the public health services, answers from the different respondent groups differed. The share of votes given to the regional health‐care politicians increased among the politicians and the administrators, but decreased among the physicians (Table 3). Instead, the physicians preferred the parliamentary politicians to have a greater say, and heads of departments were also more frequently chosen by doctors. Eighty‐seven percentage of the politicians and 63% of the physicians chose politicians on national, regional or local levels.

The answers were differently distributed when the question concerned whether it was primarily the task of the politicians to take comprehensive decisions regarding exclusion, or whether it was the duty of physicians to single out certain conditions or patients that should not be covered by public provision (Table 3). The majority of physicians had the politicians as first choice, whereas the majority of politicians considered that the physicians were most suitable to have this responsibility. On the other hand, the results also showed that almost half the politicians considered themselves most responsible, and a third of the physicians likewise.

Discussion

The customary limitations that affect attitudinal surveys, such as framing effects, uninformed respondents and external influences, apply equally to this study. As we do not have any information about non‐respondents, there is some uncertainty as to whether the outcomes are representative of the populations studied. The age distribution among the citizens corresponds to the national average, although the group 51–60 years is slightly oversized. The stratification of the decision‐makers makes controls impracticable. No information is available on the politicians' affiliations.

The questions were framed in such a way as to be easily intelligible for professional and lay people alike. Response alternatives in the financing questions may not necessarily cover all options, but they were mainly selected as the most frequent alternatives mentioned in the health‐care debate. Many respondents might have chosen to lower national spending in other fields if that alternative had been given. ⁵ The possibility of reallocating money from other public areas is normally less of an option in Sweden, as health‐care is predominantly financed by separate county‐council taxes. Therefore that response alternative was not offered.

Citizens' high expectations of public health services, and experts' more moderate attitudes, are in line with what has been found in several British and Swedish studies. ^{5 , 7 , 17 , 18} The answers from the general public could indicate a massive ignorance on health‐care limitations, or they could be interpreted as normative statements saying that this standard is what patients should be expecting. ¹⁸ Whatever the underlying motives for public support of extensive public health‐care commitment might be, the comparison with the attitudes of the experts still raises questions.

It may be that some of the differences between the public and doctors and administrators are because of socio‐economic or political differences between the groups. Another explanation is that doctors and administrators are more experienced and have far more knowledge about the costs and benefits of different interventions. This could explain why the public might tend to focus an ideal state of thing, whereas doctors' and managers' more restrictive attitudes reflect their practical insight on the unavoidable finiteness of health‐care resources. Among the decision‐makers, the biggest differences are between the physicians' and the politicians' views on joint financing. If more money is needed, increased taxes are least popular among the physicians and most popular among the politicians. The opposite goes for complementary private insurance and increased user fees. The question arises as to why this is the case.

Doctors may see complementary private funding as a way of increasing resources and giving health‐care providers a better chance to meet high expectations. It could also be a way of decreasing demand, thereby gaining more time and resources to serve the patients with greatest need. One could also expect politicians to account for their preferences for increased taxes in terms of safeguarding equity in health‐care utilization (as well as protecting their own influence). ¹⁹ The politicians' scope is different than the physicians' in its stress on access for everyone, and this could give rise to doubts as to politicians' efficiency as rationing agents. Access is often interpreted in terms of the individuals' time or money costs ²⁰ and removing financial obstacles could lead to `entrance to waiting lists' which does not necessarily mean better access.

The three expert groups are quite concordant in their views on who has the greatest influence on resource allocation in public health‐care; but they disagree on who should have the greatest influence. Physicians are much more reluctant than others to give regional health‐care politicians more influence. Instead, more of them choose politicians at national level or heads of departments.

Concerning the delicate question of excluding services from public financing, which is at the heart of explicit rationing, there is a tendency to shift responsibility on to the other group. In Sweden, this division between the political decision and clinical criteria is not at all clear, and this also seems to be the case in Great Britain. ^{13 , 14} The `pedagogic gap' between the public's high expectations of public health services and decision‐makers' more restrictive attitudes create difficulty when deciding who is accountable in the resource‐allocation process. <sup>21</sup> Differences between physicians' and politicians' opinions on financing and responsibility for prioritization make the question of accountability even more important. The effects of future prioritization efforts might be quite different, depending on how public health‐care will balance the vertical and the horizontal approaches for priority setting. Vertical priority setting (effective resource utilization and treatment of the `right' patients) is the province of physicians, while horizontal priority setting (decisions about the exclusion of certain diagnostic groups or treatments) is seen as the responsibility of politicians.

Involving patients and citizens in decision on resource allocation must be based on a deeper public understanding of the economic realities, but also on a stronger feeling of responsibility. If attitude surveys are used as instruments in horizontal priority setting, as complement to the ideally evidence‐based vertical priority setting, a quite new perspective will open up. ²² There is a tension inherent in the two approaches. While the vertical approach often emphasizes the importance of cost‐efficiency and medical optimality, the `participatory approach' puts the emphasis on the subjective, value‐based nature of public input. ¹⁴ There need not be any conflict between these two strategies if the attitudinal research is conducted in awareness of its methodological limitations. Then its legitimacy is strengthened.