Health Board Elections and Alternative Pilots: Literature Review

4 ELECTED HEALTH BODIES

4.1 This section reviews the evidence on elections to health bodies. The evidence on elected health bodies in Canadian provinces is patchy. There is more comprehensive evidence on England and New Zealand, and in these cases evidence will be summarised under the following headings:

• the elections
• the relationship between the Board and the public
• the level of influence elected members had
• evidence relating to elected Board effectiveness.

Canada: Regional Health Authorities

4.2 Canadian health services have a long history of public participation, and there is an extensive body of evidence from various periods. As part of a major programme of 'regionalisation' during the 1990s, all the Canadian provinces except Ontario devolved more decision-making power to local organisations. One of the main goals of this process was enhanced public involvement (Maddalena, 2006), and in many cases this involved considering elected Boards at regional level. However only a few provinces actually instituted elections, and all have since abandoned them for appointed Boards. Reasons given for this include the financial costs of elections and the need to ensure a particular skills mix on Boards. One commentator summarised the Canadian experience with elected Boards thus:

"On one hand, such elections will not fracture accountabilities but will increase democracy. On the other hand, elections constitute an expensive additional process that will hardly change board outcomes and, besides, 10% voter turnout is not really democracy." (Lomas, 2001, p. 356)

One 1999 Canadian study commissioned to advise on governance models for new health bodies concluded strongly:

"Preference for appointed members over elected members is overwhelming, in the specialized literature as well as in the opinions of key informants. The only question is whether appointments should be made by central authorities, by the board itself, or both ways." (Forest, Gagnon, Abelson, Turgeon, & Lamarche, 1999, p. 4)

4.3 While Canada's experiments with elected Boards are much discussed, it has been a struggle to identify primary research evidence. The research that does exist struggles to distinguish effects of election from the wider context of health reforms. Four provinces other than Saskatchewan have had elected Health Boards. However no primary research evidence was identified on these cases. New Brunswick had elected Regional Health Authorities between 2004 and 2008. In the 2004 election, 110 candidates stood for 53 positions. 74.5% of candidates, and 72% of those elected, were male. Election turnout was 47% (Office of the Municipal Electoral Officer, 2004). In 2008, after only one election, the Regional Health Authorities were merged into a smaller number of RHAs with fully appointed Boards in 2008. The decision was justified by the need for particular skills for Board membership

"In keeping with the best practices in corporate governance and the greater need for specific expertise to manage the significantly enlarged RHA organizations, the board members of the new RHAs will be selected on the basis of required skills and competencies as well as geographic, linguistic and gender considerations." (Province of New Brunswick, 2008, p. 22)

Alberta's Regional Health Authorities had 2/3 elected and 1/3 appointed members from 2001 but since 2004 all members have been appointed by the Minister (Government of Alberta, 2004). Quebec's Regional Health and Social Service Boards were elected until 2001 (Abelson et al., 2002). Prince Edward Island elected its Regional Health Authorities until their dissolution in 2005 (Elections Prince Edward Island, 2009; PEI Health Governance Advisory Council, 2009).

Saskatchewan

4.4 Saskatchewan created Regional Health Authorities in 1992 and moved to a system of partially elected Boards in 1995. Two thirds of each Board were elected on a ward basis, and one third appointed. Three elections took place, at the same time as municipal elections.

Year	Acclaimed candidates(no election due to lack of candidates)	Voter turnout	District election costs (GBP)[1]
1995	30% (69/232)	33%	477,748
1997	68% (85/125)	25%	244,439
1999	65% (82/127)	10%	242,656

Table 2: elections for Saskatchewan RHAs: candidates, turnout, costs (Saskatchewan Health, 2001, p. 59).

4.5 There were concerns about elected Boards being captured by sectional interests who would make decisions against the general good of the population (Lomas, 2001). The Saskatchewan model was unusual for its decision to neither prohibit nor discourage health care providers standing as candidates: 47% of Lewis et al's survey of Board members were or previously had worked for a health care provider (S. J. Lewis, et al., 2001). However there was no evidence that elections had particularly politicised the Boards. 25% of respondents felt that their role was most like that of a school board member, compared to 14% who felt it was most like a member of a legislature (S. J. Lewis, et al., 2001). 91% of respondents indicated that they would support a decision they believed to be right, even if it were opposed by the community, and 30% felt that their input to the Board was not strongly influenced by people in the community (S. J. Lewis, et al., 2001). A significant majority of respondents felt mostly accountable to all resident in the district (76% of respondents): but elected members were more likely than appointed members to feel most accountable to residents of their ward (S. J. Lewis, et al., 2001). Overall, Lewis et al (2001, p. 346) found

"surprisingly few differences in perception between elected and appointed members".

4.6 Frustration about a lack of Board autonomy was a notable finding. 76% of all respondents (82% of elected and 64% of appointed members) agreed that Boards were legally responsible for things over which they had insufficient control, and 64% of elected respondents agreed they had less authority than they had expected when districts were formed (S. J. Lewis, et al., 2001).

4.7 One research project on the experience of the Saskatchewan elected Boards concluded that

"Neither the worst fears nor the highest hopes have been realised" (S. J. Lewis, et al., 2001, p. 347).

In 2001 it was announced that the overall number of RHAs would be reduced, and their Boards would be fully appointed. The announcement was explained as follows:

"This system has been costly and not very popular. With few candidates coming forward for elected positions, and poor voter turnout, board elections have not proven to be an effective way to involve the public." (Saskatchewan Health, 2001, p. 59)

England: Foundation Trust Boards of Governors

4.8 Foundation Trusts (FTs) were created in 2004. They have an unusual governance structure consisting of a membership made up of local people, who then elect a Board of Governors (UK Department of Health, 2009). Policy documents described this structure as modelled on "co-operative and mutual traditions", but commentators argue there is no evidence that FTs have fulfilled this brief (Allen et al., 2012; Wright, Dempster, Keen, Allen, & Hutchings, 2011). FTs operate under 'earned autonomy', with their Board of Directors held to account by this Board of Governors, rather than the Strategic Health Authority (Dixon, Storey, & Rosete, 2010). Each Foundation Trust has discretion in how to arrange both membership and elections, resulting in considerable diversity of method and Board structure (Day & Klein, 2005). However certain statutory provisions exist (House of Commons - Health Committee, 2008) including:

• Governors appoint the Chair and non-executive members to the Board of Directors
• Governors can dismiss the Chief Executive with a 75% vote
• Boards of Governors consist of a majority of elected members (both staff and public/patient) and a minority of appointed stakeholder members (from Primary Care Trusts etc).

Ham and Hunt (2008) found considerable variation in Boards of Governors. Their study of six FTs found sizes varying from 21 to 50, with most around 30 members.

Elections

4.9 As FTs can define their own model of membership and election it is difficult to draw conclusions about the finer details of models. In research conducted as recently as 2011, FTs reported an ongoing process of development of their representative structures (Allen, et al., 2012). Day and Klein's (2005) study of the first year of FTs identified a range of options

• whether to use constituencies or not (and type - whether geographical or of interest)
• electoral system (although most used Single Transferable Vote)
• the age of eligibility of membership (from none at all to 18).

One FT chose to use an 'opt-out' model of membership - all local residents were automatically members unless they chose to opt-out - which created an unusually large membership, and an unusually low election turnout (Day & Klein, 2005).

Election turnout has fallen from 48% average in 2004, to 25% in 2011 (Monitor, Electoral Reform Research, & Member Engagement Services, 2011). One study noted that specialist FTs (such as the Royal Marsden) tend to achieve higher election turnouts (Ipsos MORI, 2008). The number of candidates per seat has fallen slightly, and the number of uncontested elections has increased from 24% to 47% between 2004 and 2011 (Monitor, et al., 2011).

In terms of the characteristics of Governors elected, Day and Klein's early study noted high numbers of retired Governors and of Governors who have at some point worked in the NHS (Day & Klein, 2005).

Public engagement

4.10 The public role of Governors is slightly unclear. In one survey, 28% of Governors who responded hadn't been involved in any 'engagement' activities (Ipsos MORI, 2008). In two studies, the question of whether Governors should hold surgeries (in the way an MP or councillor might) had arisen (Ham & Hunt, 2008; R. Lewis & Hinton, 2008). Ham and Hunt (2008) found that while in some FTs governors were holding constituency meetings, these tended to attract only small numbers of the public. In other FTs governors had not felt confident or knowledgeable enough to do so, and in some the FT had taken the view that governors should not hold surgeries (Ham & Hunt, 2008). Allen et al (2011) found evidence of informal links between governors and the public membership, such as visits to community fairs, and found that governors often saw themselves as a conduit between public and organisation (Wright, et al., 2011).

4.11 Both Lewis and Hinton (2008) and the Healthcare Commission (2005) found that the new governance arrangements had encouraged FTs to be more open to the public, including holding well-attended public meetings. One study found that FTs made significant investments of time and money and energy to engage with its membership (Allen, et al., 2012). However the Healthcare Commission also commented that too few FTs had made specific efforts to engage traditionally poorly represented groups (Healthcare Commission, 2005, p. 9), and criticised the practice of FTs moving to hold their Board of Director meetings in private (Healthcare Commission, 2005, p. 40).

4.12 Three studies highlighted the confusion caused by multiple channels of public influence (via membership, Governors, Public and Patient Involvement Forums/LINks and local authority Overview and Scrutiny Committees) (Dixon, et al., 2010; Ham & Hunt, 2008; Healthcare Commission, 2005). Ham and Hunt (2008) argued that the removal of other channels would strengthen the role of members and Governors, but reported that some interviewees within FTs saw multiple routes of public influence as an advantage.

Influence

4.13 The studies by Dixon, Storey and Rosete (2010) and Lewis and Hinton (2008) agree that Governors have not played a very influential role. Allen et al (2012) highlight very mixed views from Governors on their own influence, ranging from having more influence than expected, to feeling excluded from key business and not given access to papers. However Dixon, Storey and Rosete (2010) and Ham and Hunt (2008) agree that the statutory powers of Governors, especially around appointments and dismissal, 'protect' their status. The potentially large number of Governors on any given Board suggests that Boards of Governors are intended as advisory, not decision-making bodies (Day & Klein, 2005). Lewis and Hinton agree, and point out the challenges of evaluating an aim as modest as Boards of Directors 'listening to' their Governors (R. Lewis & Hinton, 2008).

4.14 As well as elected Governors internal influence, several studies highlight the extent to which FTs are limited by their external accountabilities (Klein, 2003). Formally, FTs are accountable not only to their Board of Governors but, in the terms adopted by Dixon, Storey and Rosete (2010)

• 'vertically' to Parliament
• 'diagonally' to both Monitor and the Care Quality Commission
• 'horizontally' to LINks and Overview and Scrutiny Committees .

The same study found that in practice, Strategic Health Authorities also continued to hold FTs to account informally, and concluded:

"Contrary to the major policy objectives of giving greater autonomy to FTs and making them more accountable to the local population, FTs continue to look up rather than down." (Dixon, et al., 2010, p. 88)

Effectiveness

4.15 The role of the Board of Governors is different from that of a more familiar Board of Directors. In their case study Lewis and Hinton found that Governors performed a range of roles, but were not involved in day-to-day management (R. Lewis & Hinton, 2008). One Board Chair categorised potential activities as guardianship, ambassadorial, statutory/constitutional and advisory (Ham & Hunt, 2008). In a later study, it was found that governors were basically effective in fulfilling their role in holding directors to account (Wright, et al., 2011). Lewis and Hinton found some disagreement between Governors and Directors over their appropriate role in decision-making, with some Governors keen to take strategic control, while others, and most Directors, preferred the Board of Governors to focus on 'patient experience' (R. Lewis & Hinton, 2008). Day and Klein's early study identified some "meddling" with operational matters such as car parking and cleanliness, and suggest that this may be inevitable given Governors' likely concern with the 'end product' of patient experience (Day & Klein, 2005). Governors have a fairly hands-off role, with most choosing not to attend meetings of the Board of Directors (Ham & Hunt, 2008). Only 20% of Governors attend 'all or most' meetings of the Board (Ipsos MORI, 2008). Elected Governors attend more meetings than appointed (stakeholder) members (Ipsos MORI, 2008).

4.16 Turning to Governors' own assessment of their effectiveness, 27% declined to answer the survey question 'What would you say have been your main achievements as a Governor?' This is interpreted by the report's authors as uncertainty (Ipsos MORI, 2008). In another question Governors listed improved communication and better engagement as the main improvements to the FT (although only 7% and 6% respectively agreed) (Ipsos MORI, 2008). However Governors were happy with the support they received from their FT. 90% of respondents felt that their Trust kept them very well or well informed and find the Board of Directors approachable (Ipsos MORI, 2008), and 77% were very or fairly satisfied with the training and induction they had been offered (Ipsos MORI, 2008).

4.17 Ham and Hunt reported some initial tension between Governors and Directors around the role of Governors in appointments and remuneration, but found that FTs tended to move past this (Ham & Hunt, 2008). Ipsos MORI reported one Chair doubting whether Governors were 'up to' the job of selecting Non-execs, considering them to be 'amateurish' (Ipsos MORI, 2008). All studies agree that the role of Chair (a dual role chairing both the Board of Governors and of Directors) is particularly crucial to the success of FT governance (Ham & Hunt, 2008).

4.18 The Healthcare Commission report also found that many Non-Executive Directors felt that the existence of elected governors relieved Directors of the responsibility of "representing the local community", and left them able to focus on strategic decision-making (Healthcare Commission, 2005, p. 42). The Audit Commission (2009, p. 2) agreed that

"the introduction of FTs has generally reinvigorated governance processes and resulted in the recruitment of non-executives with a greater knowledge of effective risk management and board challenge drawn from private sector experience."

New Zealand: District Health Boards

4.19 District Health Boards are responsible for arranging all health services for their populations, and additionally own and manage public hospitals. Direct elections to District Health Boards were instituted in 2000 (New Zealand Ministry of Health, 2010). There had previously been directly elected Area Health Boards for a spell in the 1980s (Cumming & Mays, 2002); the switches away from this model, and back again, were driven by changes of Government (Ashton, 2001; Ashton, Mays, & Devlin, 2005). District Health Boards have up to 11 members, seven of whom are elected and four appointed by the Minster of Health, with the intention of enhancing the skill base and community representation of the Board. Through this mechanism, it is ensured that at least two Board members are of Maori origin. It is stated in the legislation that although Board members are elected, their primary accountability is to the Minster for Health. The 2000 Act created additional duties for Boards to hold Board meetings in public and to consult on strategic items (Tensenbel, Cumming, Ashton, & Barnett, 2008).

Elections

4.20 The model of elections has changed across the elections, with Single Transferable Vote (STV) optional in 2001 and then compulsory from 2004 (Gauld, 2005). In the first election (2001) a high number of candidates stood (Gauld, 2011). This has dropped off significantly in the next two. Gauld states that there is no clear reason for this, but proposes disenchantment with the system, or simply a reduction in the initial excitement as potential explanations. 80% of incumbents stood again in 2004 (Gauld, 2011). Election turnout has remained fairly high, dropping to 43% in 2007 but increasing again to 49% in 2010 (Department of Internal Affairs, 2011).^[2]

	2001	2004	2007	2010
Voter turnout (%)	50	46	43	49
Candidates per seat	7.4	3.5	2.9	2.7

Table 3: New Zealand DHB elections: turnout and candidate numbers (Department of Internal Affairs, 2011)

4.21 Gauld suggests that the influence of postal voting contributes to these strong results. Gauld's 2007 post-election survey explored non-voters' rationales, and found the most common reasons were

• 'don't know' (35%)
• 'didn't know about elections' (19%)
• 'didn't receive voting papers' (12%)
• 'no interest in elections' (17%)

Interestingly, the proportion choosing 'no interest in elections' had fallen from 30% in 2004. In 2007 voters made their decisions on candidates by

• using candidate profiles (64%)
• looking for someone they know (27%)

Only 3% had 'guessed'. The qualities sought in a candidate have remained fairly stable across the three elections, with healthcare experience proving most popular (56% in 2007) and management or finance experience far less so (7% in 2007) (Gauld, 2011).

4.22 Representation of the Maori population is a major issue in New Zealand. Four members of DHBs are appointed, and this is intended in part to deal with under-representation of Maoris through elections (Gauld, 2011). Maori representation through the elections did improve after the full introduction of STV in 2004 (Barnett & Clayden, 2007). This followed a Government campaign to encourage Maori to stand as candidates and vote (Alliston & Cossar, 2006). Nonetheless the proportion of elected members of Maori ethnicity was 8% in 2004 and 2007 (Gauld, 2011). The total population identifying as Maori in the 2006 census was 14.6% (Statistics New Zealand, 2006).

Public engagement

4.23 Barnett and Clayden found that the combination of elected members and public Board meetings did prompt a cultural change towards openness (Barnett & Clayden, 2007). Although interviewees acknowledged that speaking out in public could compromise Board members, they felt this was part of learning a new way of working. Board meetings became slower moving, with the need to explain and reiterate for members of the public present (Barnett & Clayden, 2007).

4.24 Barnett and Clayden found that Boards had very variable ways of engaging with their public. Methods included a public right to speak at Board meetings, and public road shows (Barnett & Clayden, 2007). In addition, DHBs were encouraged to create special mechanisms to consult with their Maori population, including agreements with existing Maori bodies (Alliston & Cossar, 2006) (Boulton, Simonsen, Walker, Cumming, & Cunningham, 2004). However where community engagement had improved, Barnett and Clayden found no evidence that this was as a direct result of elected members (Barnett & Clayden, 2007). Gauld similarly concludes that

"the New Zealand experience … indicates that electoral mechanisms may play only a limited role in promoting participation, and could possibly counter public involvement…an elected board may be but one of multiple, parallel methods for public participation." (Gauld, 2011, p. 9)

Tenbensel et al agree, arguing that

"responsiveness to central government strategies has far outweighed the representation of local communities in decision making" (Tenbensel, Mays, & Cumming, 2011, p. 245).

Influence

4.25 Although Barnett and Clayden found some evidence of elected members having to struggle with management to gain access to strategic decision-making (Barnett & Clayden, 2007), the literature clearly suggests that the main barriers to the effective influence of elected members are constraints imposed on Boards by central Government. Despite Boards spending time on prioritisation exercises, one study found

"DHBs often lacked confidence that they could act on prioritisation even if they wanted to, because they would not get such decisions past central government and/or the local community." (Tensenbel, et al., 2008)

Gauld found that some elected members presented themselves to their constituents as mere "Government messengers" (Gauld, 2011). Barnett and Clayden similarly emphasise a lack of scope for District Health Boards to exert strategic direction (Barnett & Clayden, 2007), and Ashton discusses situations where Ministers have reversed DHB decision, undermining elected members (Ashton, 2005). Most Board members see planning as developing a local version of national strategic plans (Tensenbel, et al., 2008), and accordingly influence is more likely to be over issues around service design and delivery. The consensus seems to be that the shift to local decision-making has been outweighed by other policy trends:

"Despite the formal organisational shift to local (i.e. DHB level) decision-making, the pressure in the opposite direction to hold local agencies accountable for their use of public funds has, if anything, increased over time." (Ashton, et al., 2005)

Effectiveness

4.26 The literature identifies a steep learning curve for new elected members. Chief Executive Officers reported a lack of technical skills in some cases, with particular gaps around financial management (Barnett & Clayden, 2007). However they also stated that they valued new members' strong networks and community contacts (Barnett & Clayden, 2007). Chairs talked about the hard work put in to equip new members with the necessary skills, and Barnett and Clayden conclude that

"While Boards may not be seen to have the necessary skills, and the ability to fill those skills gaps through appointment was constrained, the notion of Board development is strongly present. When accompanied by good leadership and supportive management the capability of the Boards can clearly be raised to appropriate levels." (Barnett & Clayden, 2007)

While there is some evidence of early tensions between management and elected members, with the careful management of Chairs, effective team-working was achieved in most cases (Barnett & Clayden, 2007)

4.27 Elections to health bodies have been held in a number of systems: in some Canadian provinces in the 1990s; in New Zealand in the 1980s and again since 2000; and for Foundation Trusts in England since 2004. Voter turnout for these elections is generally disappointing, and tends to fall over time, along with numbers of candidates. The extent to which elected members engage with the public varies, but in some cases the presence of an elected Board seems to encourage organisational openness and transparency. Elected members often have a different skill-set from previously appointed members, and experience a steep learning curve. Notably, in each instance of elections, research has found that elected members are surprised and often frustrated by the lack of autonomy their boards have from central government control.