Mary Dixon-Woods is Professor of Medical Sociology and Wellcome Trust Senior Investigator at the University of Leicester, Deputy Editor of BMJ Quality & Safety, and a member of the Health Foundation’s Improvement Science Development Group. We spoke to her about why change in the NHS is so hard, and how national bodies can do more to help.

There’s a lot of talk about the need for culture change at the front line of the NHS, but I think a lot of that culture change needs to come from the very top. We’ve got into the habit of exhorting people at the sharp end of care to sort things out, without giving enough attention to what the blunt end needs to look like.

There are so many different agencies and bodies giving conflicting or confusing directions to organisations. And expectations can be unrealistic. That leads to bullying at lower levels, because people come under pressure to meet targets or comply with guidelines and they simply don’t have the capacity or capability. There needs to be more coordination and integration, and more thought given to exactly how plans can be converted into practice in how care is delivered.

I think the NHS does very well given how slim the managerial workforce is. Compared to what I see in some high-functioning US organisations, the number of people running NHS organisations is extremely small, which limits our ability to change. There just aren’t enough people doing the ‘making the trains run on time’ work. What management capacity there is gets used up in chasing external demands. Which then means there’s very little headspace available to focus on improvement, and the operational design skills aren’t necessarily there.

We shouldn’t be starting from scratch every time. We keep coming up with new ways of doing things in every new service line and organisation, when actually we should be agreeing on solutions that are broadly similar across organisations and customising where appropriate. But we don’t have the right structures in place to help organisations work together to solve problems in networks and collaboratives, rather than people trying to sort everything out by themselves. Initiatives like the 5000 fellows will help with this.

What we need is a more intelligent approach to intelligence. Measurement and targets have a very important role. But we’ve got locked into this pattern of aggressive responses to ‘poor performance’, without bringing in intelligence about how we can support that organisation to make things better. Sometimes problems are bigger than that one organisation; we need to understand that often what pops up in one organisation is being created at a much deeper level of the system as a whole.

There is a big disconnect between what it means to call for something and how that actually manifests at the front line of care. We did a big study on culture and behaviour in the NHS which showed that frontline staff feel there’s often little insight or understanding into what’s involved in making the changes dictated from on high. Something that might appear simple to implement to policy makers or senior management could turn out to be extremely difficult or frustrating at the sharp end, possibly even introducing new risks. The organisations that tend to be more successful are the ones that bridge that gap, and understand that the blunt end has a lot of responsibility for facilitating and supporting the work of the sharp end. And that blunt end includes policy makers and other senior decision makers.

Involving frontline staff more in policy and decision making could be very valuable. It would need to be done well though. We’ve had a lot of false starts with patient and public involvement. It’s really important that we design a system so that it meets the needs of staff and has the care of patients at the centre.

A lot of the problems that arise do so because we haven’t sorted them out at the level of the whole system. For example, we’ve known for 20 years that the way medicines are labelled and packaged is a risk for patient safety. But this isn’t something that any individual GP practice or hospital can sort out. In fact, it’s an international problem, not a national one. The NHS has got an opportunity it’s never quite seized to use its national status to sort some of these things out.

It’s both a challenge and an opportunity that improvement often happens at such a small scale in the NHS. Small scale improvement is creative and can engage people at a local level, but it means we’re constantly reinventing the wheel across organisations. There will always be an extremely important role for those small projects. But I think we need to think bigger about some things. It’s about fitting the solution to the problem; we have too much centralisation over some issues and not enough over others.

We could do a lot more to lever the national character of the NHS. I’ve been in the States for six months now and people keep saying to me ‘oh you’re so lucky to have a national system’. And I keep thinking, ‘yeah, but we’re not making enough of it!’ And that doesn’t mean every national solution needs to be led by the Department of Health or some other national body telling us what to do. We could be harnessing the national system to find collective solutions, and strengthening the horizontal connections between professional groups and service areas, rather than defaulting to vertical command-and-control on one hand or uncoordinated small local efforts on the other. It needs a shift in thinking.