Health Board Elections and Alternative Pilots: Literature Review

3 CONCEPTUAL OVERVIEW OF PUBLIC INVOLVEMENT

3.1 The evidence on public involvement comes from a variety of health systems and from the 1960s to the present day. Terminology varies across time and health systems. This section discussed three dimensions of public involvement. How can we categorise and what do we call the activities undertaken? Who should be involved through public involvement? What do policy-makers seek to achieve by involving the public?

What?

3.2 One study of cancer patients, cancer service staff and Health Board Chief Executives in Scotland found that involvement suggested very different activities to different people, including

• Patient satisfaction questionnaires
• Health professionals' knowledge of their patients' problems
• Patients attending group sessions for support
• In the case of Chief Executives, 'getting the message across' to the public (Forbat, Hubbard, & Kearney, 2009).

The evidence base in this area is confused by the range of overlapping and sometimes ill-defined terminology. Since the 1960s there have been attempts to step back from the policy terminology of particular innovations and offer frameworks that categorise and assess activities. One of the best known conceptual models in this area is Arnstein's (1969) 'ladder of participation', which sought to distinguish activities which offered genuine empowerment, from those which were tokenistic or manipulative. Charles and Di Maio (1993) offered a simpler model. More recently, the concept of 'co-production', defined as "active involvement of the public in policy decision and/or service design/delivery" (S. Martin, 2009) has become more popular as a goal than ideas of citizen control.

Table 1: Models of participation

3.3 More recent work has questioned the relevance of these hierarchies. Tritter and McCallum (2006, p. 165) reject Arnstein's ladder metaphor and propose instead a mosaic, arguing that

"A linear, hierarchical model of involvement… fails to capture the dynamic and evolutionary nature of user involvement. Nor does it recognise the agency of users who may seek different methods of involvement in relation to different issues and at different times. Similarly, Arnstein's model does not acknowledge the fact that some users may not wish to be involved."

Some commentators have cautioned against a 'more is better' approach, focused solely on progressing 'up the ladder'. Several empirical studies have explored patients' preferences for involvement in different types of decision-making, finding that 'citizen control' may not appeal to all, or even most, of the population (Litva, Canvin, Shepherd, Jacoby, & Gabbay, 2009; Litva et al., 2002; Thompson, 2007). Church et al (2002) argue that 'citizen governance' type roles with time-consuming responsibilities appeal to a very small section of the population. They argue that more meaningful participation can take place "outside of the boardroom". A helpful distinction in the Canadian literature is between citizen engagement (where authorities engage with the public to find out their preferences and needs) and citizen governance (where members of the public become part of the authority, taking on the work and responsibility of running a service) (Church, et al., 2002).

Who?

3.4 This literature review focuses on involving people as residents of a geographical area. Many (if not most) will be past, present or future service users, but their entitlement to be involved comes from their status as a member of the public. Public involvement activities have tended to attract participants from particular sections of the population, which can lead to concerns that only "the usual suspects" are involved (House of Commons - Health Committee, 2007). A 'catch 22' situation can arise where participants are expected to be both 'ordinary' members of the public and extraordinarily able to understand the complexities of public service management (Learmonth, Martin, & Warwick, 2009).

3.5 Such concerns are often expressed as being about representation. Pitkin's (1967) account of representation explains the distinction between different types. For present purposes the most relevant types are:

• Formal representation: representation requiring formal acts of authorisation (for example by election) and accountability (for example by regular re-election).
• Descriptive representation: representation requiring that those people representing the public exactly resemble them in (demographic) characteristics (for example, the use of a 'statistically representative sample' in a survey).
• Representation as 'acting for': representation which entails speaking and acting on behalf of a given constituency, regardless of mode of appointment or demographic characteristics (for example, an appointed employee representative on a Health Board).

Any given public involvement mechanism will involve one or more of these types of representation, but it is helpful to distinguish between making a Board more descriptively representative of the population, or strengthening formal representation by making the Board answerable to the public.

Why?

3.6 Concerns to involve the public are often particularly acute in systems with a health service funded by general taxation and provided by the state. In private or social insurance systems, patients are more likely to be able to express their preferences via 'exit' (Hirschman, 1970), simply, by taking their business elsewhere. Nonetheless it is possible to distinguish two rationales for public involvement within a National Health Service. Martin (2008) describes these as democratic, or technocratic. Technocratic rationales suggest that empowered service users make better decisions for the public service in question, particularly as health professionals are recognised as fallible, and the citizenry as increasingly well-educated and with greater access to information (Coulter, 2002). Democratic rationales assert that as a tax-funded universal public service, health services should be accountable to the citizenry, whether via their elected representatives or through specially designed public involvement. Abelson and Eyles (2004) suggest a third rationale for participation in health care specifically, which is the development of more educated (and, implicitly, more responsible) service users.

Conclusion

3.7 These debates highlight the complexity of public involvement in health bodies. There are many potential models of involvement, groups to involve, and indeed aims for involvement. There is no clearly superior option and there will be trade-offs involved in the decisions. For example, time-consuming methods might produce more accountability but exclude the time-poor, majoritarian democracy might not represent minority opinions, and broad-based 'engagement' might include much of the population but offer little empowerment. The goal of greater public participation needs to be defined with some specificity, and matched with appropriate mechanisms to avoid disappointment.