Background
Patient and public involvement (PPI) in healthcare is an increasingly important aspect of European health systems [
1‐
5]. Involvement is framed as positive for individuals, the health system, public health, as well as for communities and society as a whole [
6]. Yet, in the European context we have insufficient knowledge of who and to what extent people want to be involved in healthcare decisions, and whether this varies by individual characteristics or differs between countries. In this article we measure the wish to be involved in health care decisions in two countries with Beveridge type health systems – England and Sweden. The core principles of Beveridge systems are universality and equity (e.g. Denmark, Finland, Greece, Ireland, Italy, Norway, and Spain), in contrast to systems of Bismarck-type that are built on the principles of plurality, liberty and solidarity (e.g. Austria, Belgium, France, Germany, Netherlands, and Switzerland). Importantly, Beveridge systems are funded by means of general taxation while Bismarck systems are funded by means of earmarked premiums, mainly from salaried employees [
7,
8]. In these two countries, we investigate preferences for involvement in two types of decisions or activities; individual treatment and local service development. Individual treatment decisions are linked to the perspective of the health service user (the patient) focusing on decisions about one’s own care. Strategic decisions about health services and policy at local or national level are associated with the perspective of a
public policy agent (member of the public) and take account of a broader public interest [
4,
9].
Earlier research elucidating preferences for involvement in healthcare has focused on the health service user’s preferences for involvement in treatment decisions, and shown significant variability in how much patients want to participate in individual decisions [
10‐
12]. For example, it has been shown that 69% of patients with long-term conditions preferred to leave their medical decisions to their physicians [
13] and that as few as 1.2% wanted an autonomous role [
14]. Thus, evidence in the literature is inconclusive leading Nota et al. to conclude that “it remains difficult to explain or predict patient preferences” [
15]. Most studies investigate involvement preferences in restricted patient populations rather than in the general population; one exception being Levinson et al. [
16]. From a population-based study in the U.S they concluded that 52% preferred the physician make the decisions (characterised as a passive patient role by Arora and McHorney [
13]). Women, more educated and healthier individuals were less likely to prefer the physician to make the decisions. Up to the age of 45 years the preferences for an active role increased.
There has been far less research on individual preferences for involvement in strategic decisions about health services and policy; the role fulfilled by the public policy agent. An annual survey in one of the 21 Swedish regions found that 64% of the residents thought the public should participate in prioritization discussions [
17]. Similarly, in the UK, a focus group study showed that people expressed a strong desire for the public to be involved in different types of rationing decisions, but evidence of the willingness to be involved at individual level is much less explicit [
18]. It may be that people are more interested in being involved in relation to some types of healthcare intervention such as organ donation or end of life care [
19]. Anecdotal evidence from public involvement activities suggest that those individuals who actually get involved are “the usual suspects” [
20] or the “vocal majority” [
21]. It is not clear if the usual suspects are also more active in individual treatment decisions.
To investigate preferences, we gathered and compared data on the preferences of the general population in Sweden and England for different aspects of patient as well as public involvement. Specifically, (1) we sought to compare and contrast expressed preferences for individual decision-making in a medical setting with the desire to be involved collectively in the organisation and provision of local health care services in England and Sweden. Secondly (2) we sought to understand how the orientation towards acting locally was informed by expectations of such involvement leading to change, and finally (3) we sought to explore the implication of individual sociodemographic characteristics (gender, age and educational attainment) on reported preferences. No previous studies have investigated the desire to be involved in individual treatment decisions (the health service user perspective) and in local decision making concerning the organisation and provision of health and social care services (the public policy agent perspective) in the same individuals in a general population. The combination of these perspectives contributes to a more comprehensive understanding of the conditions for involvement in health; in individual health matters as well as matters of public concern. Encouraging patients to take more control when they are ill may prove to be an effective tool for improving public health [
22] engendering the co-production of wellbeing, and empowering citizens and putting them at the “heart of the system” is one of the four core principles of the EU Health Strategy [
23,
24]. For instance, in Sweden, participation and influence in society is seen as one of the fundamental conditions for public health. Public health policy in Sweden points out that a lack of influence or lack of opportunities for involvement in decisions that affect people’s living conditions and the development of societal functions are negatively correlated to health [
25]. What is less clear is whether there is a link between the interest and likelihood of individuals to be involved in personal treatment decisions and to be involved in the development of local health services.
We chose to compare Sweden and England, countries which have similar health systems of Beveridge type. The English and Swedish health systems differ, however, in relation to the degree of decentralization and sub-national democratic governance, degree of patient centeredness and the set-up for PPI [
26,
27]. Health system reform over the last 20 years in both countries has promoted markets and competition underpinned by a commitment to patient choice [
28]. In Sweden, choice of provider in primary care (backed up by freedom of establishment for private providers) has been the central mechanism for strengthening the patient’s position since the mid-2000s [
29,
30], from 2015 backed by the first patient law promoting the patient’s position, integrity, autonomy and participation (Patientlag 2014:821).
Although patient choice has also been at the top of the policy agenda in England [
28], it is less apparent in practice as efforts to retain people’s confidence in the services provided have combined an economically motivated consumerist approach aiming at improving efficiency with ideologies of democratic public engagement [
31]. Despite significant development of opportunities for involving both patients and the public through voice as well as choice mechanisms and to shape local health priorities under successive legislation in 2001 and 2007, these were changed by the Health and Social Care Act 2012. People are currently involved through local
Healthwatch organizations (the consumer champion for health and social care [
32]), the
NHS Citizen (a national programme to give the public a say on healthcare matters and influence NHS England decision making [
33]) or through local
patient participation groups (PPGs); the latter a contractual requirement for all English primary care practices from 2015 which should also make reasonable efforts for PPGs to be representative of the practice population [
34]. NHS England– that set the priorities and direction for the NHS – are also working to transform participation in health and care and help commissioners of health and care services to involve patients and carers in decisions relating to care and treatment and the public in commissioning processes and decisions. To involve the public, commissioners should, for example, develop joint approaches with local authorities, health and well-being boards, local Healthwatch, voluntary groups and other organisations, especially those who have existing relationships with local communities [
35].
In comparison to England, public involvement in Sweden is more embedded in democratic institutions; the public primarily being represented by democratically elected representatives in the 21 healthcare regions, implying that healthcare decision-making is based on public interest [
36]. Public involvement is generally framed as a way to influence the political process and political decisions within a region. To increase people’s opportunities to be involved between regional elections (between-election democracy) and in more practical aspects of service planning, more participatory forms of public involvement have gradually emerged gathered under the umbrella term “citizen dialogue” [
37], which refers to mechanisms such as citizen surveys, panels or public meetings. Both Sweden and England have a long history of involving both patients and the public within similar Beveridgian health systems and their comparison provides an opportunity to investigate the patterns in preferences for involvement.
Results
Together, the two samples included 3125 respondents, 52% of whom were English and 51% of whom were female. The average age was 51 years (SD 20). The proportion of respondents with a low level of education was 34%, medium level education 32% and high level of education 34%. Percentages for Q1-Q3 for both countries together are presented in Table
3.
Table 3
Percentages for Q1-Q3 in England and Sweden
1 | 1485 (48) | 577 (19) | 339 (11) |
2 | 547 (18) | 773 (24) | 773 (25) |
3 | 620 (20) | 828 (27) | 1049 (34) |
4 | 151 (5) | 512 (16) | 536 (17) |
5 | 191 (6) | 386 (12) | 247 (8) |
6 (Don’t know) | 131 (4) | 84 (3) | 84 (6) |
N | 3125 | 3125 | 3125 |
Descriptive results
The distribution of Q1-Q3 when comparing Sweden and England showed a higher proportion of the population in Sweden agreeing to all three questions. In terms of willingness to make decisions about their own medical treatment (Q1) the two countries had similar proportions of respondents being neutral while respondents from Sweden were less neutral than the English respondents with regards to willingness to be involved in local decisions concerning the organisation and provision of health and social care services, as well as regarding people’s ability to help make improvements to health and social care services (Q2 and Q3). Respondents from the English sample showed a higher proportion that disagreed across Q1 and Q2, with overlapping confidence intervals with regards to Q1 (see Table
4).
Table 4
Descriptive presentations Q1-Q3
Agree (%) | 66 ± 2.4 | 70 ± 2.4 | 33 ± 2.3 | 55 ± 2.6 | 36 ± 2.4 | 39 ± 2.6 |
N | 1005 | 1027 | 531 | 788 | 574 | 538 |
Neutral | 21 ± 2.0 | 20 ± 2.1 | 30 ± 2.2 | 25 ± 2.2 | 40 ± 2.4 | 31 ± 2.5 |
N | 319 | 301 | 475 | 353 | 626 | 423 |
Disagree (%) | 13 ± 1.7 | 10 ± 1.5 | 37 ± 2.4 | 21 ± 2.1 | 24 ± 2.1 | 30 ± 2.4 |
N | 197 | 145 | 595 | 299 | 377 | 406 |
QI – Patient participation in treatment decisions
Excluding the don’t know-answers (1.7% in Sweden and 6.1% in England) a χ2-test suggests that there was an overall country difference in the preferred level of involvement in treatment decisions (χ2 = 22.6479, df (2), p = 0.001). In England 66% (± 2.4) answered that they want a health professional such as a GP or a consultant to make decisions about their treatment, compared to 70% (± 2.4) in Sweden. Furthermore, 13% (± 1.7) of the English sample responded that they want to make their own decisions and not rely on a health professional, compared to 10% (± 1.5) in Sweden.
Q2 – People’s wish to be involved in organizational decision-making
Excluding the don’t know-answers (3.4% in Sweden and 1.5% in England) a χ2-test shows that there was a country difference in people’s wish to be involved in organizational decision-making (χ2 = 146.9667, df (2), p = <0.001). The Swedes were more positive towards being involved as 55% (± 2.6) agreed with the statement that they would like to be involved in local decision making concerning the organisation and provision of health and social care services compared to 33.0% (± 2.6) in England. Furthermore, 21% (± 2.1) of the Swedish disagreed compared to 37% (± 2.4) among the English.
Q3 – Are people able to help improve services?
Excluding the don’t know-answers (9.5% in Sweden and 2.9% in England), a χ2-test suggests that there was an overall country difference in the perception of whether people are able to help make improvements to health and social care services at the local level (χ2 = 14.4990, df(2), p = 0.0067). The Swedes to a greater extent agreed with the statement that people are able to help make improvements (39% ± 2.6, compared to 36% ± 2.4 in England) as well as disagreed (30% ± 2.4, compared to 24% ± 2.1 in England).
Country differences and sociodemographic characteristics
Controlling for gender, age and education, there were significant differences between England and Sweden regarding involvement in treatment decisions, as well as in regard to decisions about the organisation and provision of health and social care services (see Table
5). There were no significant differences between the English and the Swedish respondents in relation to whether people were able to help make improvements to local health and social care services.
Table 5
Country differences and sociodemographic characteristics
Agree (1 + 2) | Age | .997 (.003) | 1.015*** (.004) | 1.007 (.004) |
Female | .787* (.088) | 1.294* (.156) | .905 (.106) |
Low level of education | 1.272 (.193) | .518** (.103) | .931 (.175) |
High level of education | 1.013 (.134) | .998 (.135) | .881 (.115) |
England | .697** (.084) | .642** (.084) | .943 (.126) |
Q1 | NA | 1.285** (.101) | 1.018 (.079) |
Q2 | 1.043 (.073) | NA | .684*** (.044) |
Q3 | 1.199 (.081) | .830** (.057) | NA |
Constant | 3.314*** (.833) | .979 (.283) | 1.674 (.490) |
Disagree (4 + 5) | Age | 1.004 (.004) | 1.017*** (.004) | 1.014*** (.004) |
Female | .961 (.156) | 1.144 (.147) | .964 (.116) |
Low level of education | .842 (.164) | .773 (.148) | .674* (.128) |
High level of education | .865 (.168) | .833 (.128) | 1.217 (.166) |
England | 1.616 ** (.286) | 1.884*** (.275) | .884 (.116) |
Q1 | NA | .889 (.083) | 1.000 (.081) |
Q2 | 1.051 (.106) | NA | .920 (.062) |
Q3 | .805* (.088) | 1.132 (.082) | NA |
Constant | .512 (.220) | .313*** (.095) | .499* (.149) |
| Number of observations | 2782 | 2782 | 2782 |
| Pseudo R2 | .0173 | .0479 | .0184 |
In England, people were less likely to want a health professional such as a GP or hospital doctor (consultant) to make decisions about their treatment (0.697**) and more likely to want to make their own decisions (1.616**). Women were less likely than men to want a health professional to make decisions about their treatment (0.787*). People who disagreed more with the statement they wanted a health professional such as a GP or consultant to make decisions about their treatment (that is to a greater extent wanted to make their own decisions) were also more likely to state that they would like to be involved in organizational decision-making (1.285**). We found no significant effect of age or education. Controlling for sex, age and education, there were no significant differences between Swedish and English respondents regarding whether people were able to help make improvements in the local health service (0.943 ns). People reporting lower educational attainment were less likely to disagree with the statement (0.674*) whereas older people were more likely to disagree (1.015***). Furthermore, people who disagreed more with the statement that people are able to help make improvements were less inclined to want to be involved in organizational decision-making (0.830**).
The most pronounced country differences related to whether people wanted to be involved in decision making concerning the organisation and provision of health and social care services in the local area. The English respondents were less likely to agree with the statement (0.642**) and more likely to disagree (1.884***). Women were also more likely to want to be involved (1.294*), and the wish to be involved increased slightly with age (1.015***). However, the desire to be involved also decreased slightly with age (1.017***), and those with lower educational attainment were less likely to want to be involved (0.518**). Furthermore, people who disagreed more with the statement that they wanted to be involved (that is wanted less involvement) also believed less in people’s ability to help make in improvements in the local health service (0.684***). A reduced likelihood of wanting to make their own treatment decisions was also found among those who did not want to be involved in local service improvement (0.805*). This means that those who did not want to make individual treatment decisions also did not want to be involved in decision making about the organisation and provision of health and social care services.
Conclusions
There are complex patterns behind preferences for involvement and there is still little theorising on the motivation of service users to get involved [
55]. Views on involvement may vary between cultures and settings [
12]. In the decentralized Swedish health system, people expressed greater willingness, compared to respondents in England, to be involved in decision making regarding the organisation and provision of health and social care services in the local area; i.e. to be involved as public policy agents in decisions of a broader public interest. To what extent these preferential differences translate into differences in actual involvement and in the impact of involvement needs further exploration. In both countries, involvement policies that alter the balance between the system, the health professionals and patients (c.f. [
47]) are needed as many preferred a passive role in individual decisions. Regardless of the expressed willingness to participate and the motivations that shape involvement preferences and behaviour, "patient participation in decision-making is justified on humane grounds alone" [
57]. As essential is the right of both patients and citizens to be involved in service design and development that shape local health and social care priorities. More explicit evidence of the impact of involvement and greater attention to equity in the opportunities and take up of PPI are necessary to achieve both individual level benefits such as improved health and collective level benefits such as local community responsiveness and improved public health. Those not taking part in involvement activities are often those who have the most to gain from involvement in healthcare decision-making, that is those with greater healthcare needs [
47]. It is thus important to recruit and provide a range of opportunities to participate to involve so called non-participants [
55]. National PPI strategies that take account of the specific structural features of a health system, such as level of (de)centralization are also fundamental to increase the effectiveness of involvement rather than simply appropriating approaches used in other countries. To design health systems to promote health and well-being is integral to public health and effective PPI is an essential element of such systems [
58].