BMJ  2003;327:1421-1424 (20 December), doi:10.1136/bmj.327.7429.1421

It's good to talk

Would the NHS benefit from a single, identifiable leader? An email conversation

¹ Institute for Healthcare Improvement, Boston, MA 02215, USA, ² Strategy Unit, Department of Health, London SW1A 2NS, ³ BMJ, London WC1H 9JR

Would an identifiable leader of the NHS solve all of its problems alone or overnight—or is that magical thinking?

The NHS suffers from having no single, identifiable leader

Richard Smith, editor of the BMJ, writes: Britain's national health service, one of the world's largest public sector organisations, is highly unusual in not having a leader. Most organisations—and certainly all major corporations—have a leader. Tony Blair is the leader of Britain. Greg Dyke leads the BBC, and that huge organisation, which is comparable in some ways to the NHS, has changed dramatically since Dyke took over from John Birt. One of the best ways to change an organisation is to change its leader. This is not a cult of personality. But it does seem to be important—perhaps simply because people prefer people to abstractions—to have somebody who embodies power and accountability.

So who is the leader of the NHS? One immediate problem is that there are four NHSs—in England, Scotland, Wales, and Northern Ireland. Who then is the leader in England? Is it Sir Nigel Crisp, the chief executive of the NHS in England and also permanent secretary in the Department of Health? If he is the leader, then he's exercising a uniquely anonymous form of leadership. I doubt if one doctor in 50 could identify him.

If there is a leader of the NHS in England then it's the secretary of state for health. Unfortunately, however, secretaries of state feel like the leaders of an occupying power. The relationship between a doctor in the NHS and the secretary of state is like the relationship between an Indian villager and the viceroy of India in the 19th century: these are far away people with different sets of values. Few doctors, I suggest, would regard the secretary of state as the leader of the NHS.

We need lots of leaders at all levels in the NHS, not a single leader

Chris Ham, professor of health policy and management, on secondment as director of the strategy unit at the Department of Health, responds: You make some good points. The big difficulty in the NHS is how to achieve followership as well as leadership.

The Indian villagers don't usually want to be led. They want to be independent. Some people (not you) persist in misconceiving the NHS as a "machine bureaucracy," whereas it is a collection of professional bureaucracies. These professional bureaucracies are often self managed and resist control by non-professionals (whether John Reed [secretary of state for health], Nigel Crisp, or whoever). That is why politicians feel so much frustration.

My reading is that what we need are lots of leaders at all levels of the NHS and not just a single leader. It is leadership at the village level that is just as important as leadership at the country level. The villagers are more likely to follow one of their own than they are to follow a distant ruler. This means we need to put much more emphasis on clinical leadership and its development. Supporting clinicians so they can be followers must be part and parcel of this.

I wrote about this agenda recently in the Lancet, partly reflecting on my experience in the Department of Health. "There is now," I concluded, "a growing body of evidence on how to improve health services. Alongside the mainly negative evidence on the effects of radical solutions promulgated by health reformers, there is increasing understanding of the conditions that need to be in place for change to happen. In professional organisations... these conditions include [the] development and strengthening of clinical leadership... The trick that has to be accomplished [to achieve improvement] is to harness the energies of clinicians and reformers... The importance of linking top down and bottom up approaches to performance improvement has never been greater. On this link, nothing less than the future of organised health care systems depends."¹

We certainly need clinical leaders, but don't we need an overall leader as well?

Richard Smith: I very much agree with your general thrust that clinical leaders are hugely important and the change within the health service will come primarily from them. Where I'm much less sure is over the question of whether an organisation like the NHS needs an overall leader. It seems to me that it might. Can you think of another major institution that doesn't have a clear overall leader and yet functions effectively? I can't offhand.

I've long been interested in leadership, and one of the things I've found bothersome is that it seems to revolve around individuals. The prime task of a leader is to find highly capable individuals and to motivate them. I thus wonder why it might need to be an individual rather than a group of individuals—but I think that it might. It may, I suspect, be a very human thing. People find it easy to identify with individuals and difficult to identify with groups of individuals.

The contrast between the BBC and the NHS does seem interesting to me. Both are large, complex organisations that include many different professional and creative people, and both embody Britishness. The BBC has had difficult times, but it does generally seem to have flourished, whereas the NHS has not. Might this be something to do with having an overall leader? Certainly John Birt seemed to change the organisation dramatically, and Greg Dyke has done so again since taking over from John Birt. I'm not at all sure that I'm right with all of this, but I'm interested to pursue the thinking further. One thing I'd like to do with your permission is to share our correspondence with Don Berwick and ask for his response. Might I do that?

We may not be far apart

Chris Ham: Thanks for this. I don't think we are far apart. Of course, the NHS needs a leader as well as lots of clinical leaders. The issue I was reacting to was the view that great men (and women) are what really make a difference. Increasingly, it seems to me that we need to emphasise two things: the importance of collective leadership and not just individual leadership, and the need for leadership at all levels (especially in a big complex organisation like the NHS). We should probably also emphasise continuity in leadership: something the NHS is not very good at, and which will always be difficult when ministers come and go. I am happy for you to share with Don.

Richard Smith is guilty of magical thinking: leadership problems in the NHS are not at the top

Don Berwick, paediatrician and president of the Institute for Healthcare Improvement, writes: Thanks for including me in this dialogue. I had, indeed, seen Richard's "Editor's Choice" essay,² and, like Chris, felt it to be somewhat "over the top" in its reading of both the gap and the possible solution.

Please remember that I am an outsider and can easily develop naive views, uninformed by truly detailed knowledge of the actors and context, but, since Richard asked, here are some of my views.

I personally believe that the NHS has had superb leadership, at least as long as I have had a window on the people there. This includes both the NHS Executive team members I have known, and many of their direct reports. Alan Langlands [the immediate past chief executive of the NHS in England] was, in my view, a master manager, and Nigel Crisp, with a very different style, is also very capable. I can think of many of their deputies whom I would be proud to work for. Controversially, perhaps, I also developed great respect for many of the political leaders. For example, Alan Milburn showed a lot of courage in his work to get movement into the system, and Simon Stevens [Downing Street special adviser on health] is one of the most thoughtful political analysts and balanced decision makers I have ever known. I have met a number of very capable trust chief executives—in both primary care and acute care trusts. Liam Donaldson [chief medical officer in England] provides a world class example of trustworthy, mature, and progressive leadership in very stormy seas.

I also have to reserve a few words of praise for Tony Blair, whose focus and clarity with respect to what the NHS must accomplish are superb, and of immense (though uncomfortable) value. He has been as good in this role—setting and pursuing clear, bold aims—as many of the best corporate executives I have seen.

I may misread Richard's critique, but I believe, first of all, that interpreting the NHS leaders as a group over the past decade as anything other than remarkably competent is incorrect. The NHS is a fantastically complex organization, and wishful thinkers might hope that its troubles could be corrected by an emerging, as yet unfound, Alexander, Napoleon, or Churchill. That is magical thinking. The people actually struggling with the task of managing and improving the behemoth are about as good as you can get.

Now, Richard seems to emphasise how important the role of a single, capable senior executive leader could be to helping the NHS, and how this is missing. He seems to feel, if I read it right, that "no one is in charge" and someone needs to be. Again, I think this view may be largely based on wishful thinking. The complexity of the NHS will make it true that "leadership" will have to be a system, involving the coordinated energies of a number of top level people who should act as a team—if not in unison, at least in coordination—to get aims accomplished. The best I could hope for is a "team at the top" not a dictator at the top. Since that team will have to include a balance of political leaders and managers (because the public owns the system through its government), there will be tension, much as there often is between boards and executives. The best managers and political leaders will solve that as they best can through dialogue, respect, and lots of interaction—not by one of their number emerging as "the boss."

I completely agree with Chris's further point that the key leadership issues that the NHS really has to tackle have far more to do with local leadership capacities (and local skills in basic management) in 400 primary care trusts, 400-500 acute care trusts, and lots of other local entities with rather large budgets and influence. When thinking about the future of the NHS in the decade ahead, I lose far more sleep worrying about local leadership capacity than about leadership at the top. A primary care trust, for example, is really quite a large "company," which would require a high order of skill and expertise in its executive team. The NHS needs a very, very good plan for making sure that trust management systems are robust, reliable, and highly skilled.

Finally, the other major issue in leadership in the NHS is that of clinical leadership: the royal colleges, BMA, and such. As in the United States, a number of the important groups of this type have not yet, in my opinion, properly shouldered their obligations to participate as citizens in systemic change and improvement, rather than to act primarily as critics, guilds, or watchdogs. I do not know how to get more of those leaders into the mindset of "teammates" with the government and the NHS Executive, but, if teamness were to be evidenced among these professional leadership groups, the speed of improvement of the NHS would triple.

I still think that a single, identifiable leader could be helpful

Richard Smith: Thank you, Don, for your very useful, interesting, and compelling contribution to this debate. We probably agree more than we disagree, but I don't for an instant think that an identifiable leader of the NHS would solve all of its problems alone or overnight. That is clearly magical thinking.

Nevertheless, I continue to think that there may be extra benefits from having an identifiable leader. Every country has an identifiable leader—and so do most small, medium sized, and large organisations. Can you think of another large organisation that doesn't have an identifiable leader? Nobody has answered that question in this correspondence.

Another question I have is whether you can be an effective leader if you are largely unknown. I suspect that very few people working in the NHS could identify Nigel Crisp, whereas I suspect that every employee of the BBC knows of Greg Dyke—as do many of the customers of the BBC.

Isn't it also true that effective leaders can change large organisations in a comparatively short period of time? Again, both John Birt and Greg Dyke have done that for the BBC. It too is a large, complex, multifaceted organisation. I don't see any similar example of rapid and effective change in the NHS.

Indeed, I must confess that I don't share your optimism about the NHS. You mainly meet its leaders, most of whom, I willingly acknowledge, are charming, competent, hard working, and committed. I spend more time encountering the workforce, many of whom are deeply unhappy, and the customers, many of whom have bad experiences.

The NHS has problems but we don't need a "great man" to lead it

Chris Ham: Let me restate and hopefully clarify my position in this debate. I agree with Richard's starting point—that is, the NHS needs clear and visible leadership at the top. My initial reaction to Richard was to say that this was necessary but not sufficient. Specifically, I feel that leadership from the top must be linked with the development of leadership at all levels. And the more I have thought about this, the more important it seems to me that we develop effective clinical leadership in both primary care and hospitals.

As far as leadership at the top is concerned, I would also question Richard's emphasis on the "great men" like Birt and Dyke. We need to move beyond arguments for heroic leadership and think much more about leadership teams or coalitions. One of the great successes in the process of developing the NHS plan was that a broadly based leadership coalition was established. The disappointment is that it was not sustained. Like Don, I feel the NHS has good leadership at the top, and that the real gap is within the NHS itself.

On some of the other points, I have been very impressed by what Ken Kizer achieved at the Veterans Health Administration. When people argue that big public service organisations are impossible to turn around, I cite the Veterans Health Administration as an example of what can be done. The VHA might be used to support Richard's thesis, of course, although my sense is that Kizer did lots of things to engage his staff in the process of change, including strengthening local leadership.

I agree with Richard that staff in the NHS remain unhappy, but we need to put this in perspective. Enoch Powell wrote of his experience as a health minister in the 1960s: "One of the most striking features of the NHS is the continual deafening chorus of complaint which rises day and night from every part of it... The universal Exchequer financing of the service endows everyone providing as well as using it with a vested interest in denigrating it, so that it presents what must be a unique spectacle of an undertaking that is run down by everyone engaged in it."³

I also agree that some of the customers have bad experiences and that a lot remains to be done to change this. Nevertheless, performance is improving (as the Audit Commission, the Commission on Health Improvement, and others have argued), and I don't think the NHS is too far away from where you would expect in year 3 of a 10 year process (wasn't it Rosabeth Moss Kanter [editor of the Harvard Business Review] who said that halfway through major transformations everything feels like a failure?)

The best results come not from charismatic leaders but those who are less visible, more persistent, hard working leaders

Don Berwick: Charismatic, memorable leaders excite us, but I have the strong impression that the best results in complex organisations tend to come under the guidance of rather less visible, more persistent, hard working, "stick-to-the-knitting" leaders. So, Richard's comment that "very few people in the NHS could identify Nigel Crisp" (whether true or not) disturbs me not at all. The hardest work of effective leadership happens in the background, not on the podium, and builds support and capability for many, not fame for a few.

Lao Tsu, in 700 bc, advised leaders (so they say), "Go to people, live with them, love them, learn from them. Start with what they know, build with what they have, and work with the best leaders, so when the work is done, people can say, `We did this ourselves.' "

As for my optimism about the NHS—sorry, Richard, I simply can't help it. I do meet many leaders, in whom I have great faith, but I also meet many of the dedicated staff (some unhappy, some not) and patients (some displeased, some not) almost all of whom continue to be incredibly loyal to a pathfinding system of care that the United Kingdom started a half century ago. They are smart enough to know a good thing when they have it, even while they demand that it get better. I would trade the UK's NHS, warts and all, current leaders in the bargain, for my nation's healthcare chaos any day.

Summary points The NHS does not seem to have a single, identifiable leader, and perhaps it should in order to hasten the pace of improvement The problem of the NHS is less with leadership at the top and more with leadership at a local level But don't all major organisations have an identifiable leader? Wanting a heroic leader is magical thinking Leaders don't have to be visible to be effective

---

Do you think that the NHS would benefit from a single, identifiable leader? Vote on bmj.com, and if you do believe in a single, identifiable leader you might want to suggest one

Competing interests: DMB is the president and CEO of the Institute for Healthcare Improvement, which has had several agreements and contracts with the NHS over the past several years to help progress the modernisation plan. CH is on secondment to the Department of Health and in his work at the University of Birmingham has been involved in leadership development programmes for clinicians and others. The full competing interests of RS can be accessed at http://bmj.bmjjournals.com/aboutsite/comp_editorial.shtm

References

1. Ham C. Improving the performance of health services: the role of clinical leadership. Lancet 2003;361: 1978-80.[CrossRef][ISI][Medline]
2. Smith R. Editor's choice: Changing the "leadership" of the NHS. BMJ 2003; 326 (21 June). doi:10.1136/bmj.326.7403.0-g[Free Full Text]
3. Powell JE. A new look at medicine. Tunbridge Wells: Pitman Medical Publishing, 1976: 83.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Rapid Responses:

Read all Rapid Responses

world-wide organisation with no discernible leader

J Lindley Owen
bmj.com, 19 Dec 2003 [Full text]

Why the patients of the NHS have no leader

Jeremy G Jones
bmj.com, 21 Dec 2003 [Full text]

Is it possible for a leader to lead the NHS?

David Zitner
bmj.com, 22 Dec 2003 [Full text]

Don is ill - he needs a ROTS and RETDAM

Nigel Dudley
bmj.com, 22 Dec 2003 [Full text]

An identifiable leader for the NHS.

Morris P Doublet-Stewart
bmj.com, 22 Dec 2003 [Full text]

A single voice

Chris Smith
bmj.com, 23 Dec 2003 [Full text]

A Pro-Active Health Expert Patient Body

Robert J Cremona
bmj.com, 26 Dec 2003 [Full text]

Will NHS benefit from a single leader?

Sandip Raha
bmj.com, 1 Jan 2004 [Full text]