In this issue, Lewis and colleagues² rise to the challenge of distilling generalizations from a survey of Saskatchewan's 30 district health boards (page 343). They answer 2 policy concerns that have preoccupied the 9 provinces with devolved authorities (Ontario being the control group in this natural experiment): What happens if you introduce elections? (Saskatchewan is the only province in which board members are elected directly) and What is the impact of allowing health care providers to sit as board members? (at the time of the survey, in 1997, nearly half of the board members in Saskatchewan were current or former health care providers).

The answer to the question about the impact of elections appears to be "Not much." The major effect was that, despite the fact that three-quarters of board members felt accountable to all residents in their district, "elected members were much more likely to feel most accountable to their ward residents (17% v. 3%, p < 0.01)." This finding recalls a concern expressed in my earlier survey¹ that elections might mean "a single accountability to the community would be disaggregated into multiple accountabilities to a variety of group and geographic interests ... the 'management by interest group' approach that devolved authority was designed to overcome" (p. 819). In practice, however, Lewis and colleagues found little evidence of differences in board actions resulting from a small minority of board members adopting a ward perspective.

Add to this the fact that in Saskatchewan's 1999 board elections, voter turnout was an anemic 10%, and you have a policy choice whose potential bark proves far worse than its actual bite. No doubt the many provinces that originally promised elected boards, but have since thought better of the idea, can now use these results to either justify having elections or not. 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.

Allowing health care providers to sit on the boards may not have been quite so benign a change, although in the context of physicians on US hospital boards there is some evidence of a positive financial benefit.³ In the case of Saskatchewan's district health boards, 96% of the members who had experience as health care providers were not physicians, and they were less likely than others on the board to think that their provider colleagues supported and respected board decisions. Health care providers on the boards overwhelmingly believed that this group should have more say in running the health care system, whereas only about half of nonprovider board members believed in increasing the influence of health care providers.

One of the major challenges faced by regional boards is to confront provider interests, such as pay levels, working conditions and work location, when they conflict with community or provincial government objectives. Putting health care providers on boards may add opportunities for them to resist change when such inevitable confrontations arise. Experience to date indicates that larger regions, such as those for provincial capitals and other densely populated areas, may be better equipped to resist such incursions.^4,5 Smaller regions, with fewer resources and with boards whose members are more personally connected to their communities, are less able to resist such capture by provider interests. For this and other reasons it may be time to consolidate some of the boards into fewer and more substantive entities, better able to resist "capture" and with a broader and more significant array of complementary services available for rationalization and trade-offs.

Many regions have achieved significant gains in efficiency and effectiveness by trading off such things as hospital care for community or nursing home services or by reducing interhospital rivalry and duplication. However, further significant gains are likely to come only from opening up the 2 remaining major resource responsibilities excluded from the current scope of the regional boards — pharmaceuticals and physicians' services — to equivalent substitution and rationalization. These additions to regional authority will only occur, if they occur at all, when provincial governments are confident of the size and abilities of larger, consolidated regional boards.

The real test for the future role of regional boards in Canada is just about to begin. These boards were created in an era of fiscal restraint, at least partially to absorb and deflect blame from provincial governments for the tough choices that came with resource famine. Almost 50% of board members in both this survey and my earlier one recognized this provincial motivation. Circumstances have changed. We are now heading into an era of re-investment, with more of a resource feast than a famine in health care. There is credit to claim, not blame to diffuse.