Imagine two assembly lines, monitored by two foremen. Foreman 1 walks the line, watching carefully. “I can see you all,” he warns. “I have the means to measure your work, and I will do so. I will find those among you who are unprepared or unwilling to do your jobs, and when I do there will be consequences.

There are many workers available for these jobs, and you can be replaced.”

Foreman 2 walks a different line, and he too watches.

“I am here to help you if I can,” he says. “We are in this together for the long haul. You and I have a common interest in a job well done. I know that most of you are trying very hard, but sometimes things can go wrong.

My job is to notice opportunities for improvement – skills that could be shared, lessons from the past, or experiments to try together – and to give you the means to do your work even better than you do now. I want to help the average ones among you, not just the exceptional few at either end of the spectrum of competence.”

Which line works better? Which is more likely to do the job well in the long run? Where would you rather work?

Donald Berwick. N Engl J Med 1989;320:53–56. Whenever I’m asked what continuous quality improvement (CQI) means, this passage comes to mind. They are the opening paragraphs of the landmark

paper in which Donald Berwick introduced the concept of continuous improvement as an

ideal in healthcare just over 26 years ago.

Berwick went on to help create the Institute for Healthcare Improvement in the United States and champion

CQI throughout the world. But what do we mean by quality in healthcare?

In reality, there is no single definition, since healthcare is complex and quality has many dimensions. Viewed from the perspective of the individual and wider society, healthcare should be safe, effective, person centred, timely, efficient and equitable and performance should be assessed against all these dimensions.

It is unsurprising that when Prime Minister David Cameron needed someone to head a review into patient safety in the NHS in the wake of the Francis inquiry, he turned to Donald Berwick. His report, A promise to learn – a commitment to act – Improving the safety of patients in England, received considerable attention within the mainstream media at the time of its publication in 2013, but relatively few healthcare professionals are familiar with its contents. The key recommendations of the report are wide ranging and, if implemented, would result in profound cultural change throughout the NHS.

They are based on decades of experience within a wide range of industries and are fundamentally about creating organisations that are continuously learning and improving the service they offer.

But what forms the culture of an organisation and how can it be changed? Culture is complex and is formed by the interaction of the formal policies, procedures and values of an organisation and the people who interact with it. It is not homogeneous within large organisations, and each department or staff group will have its own sub-culture. The clarion call following the Francis report was for cultural change in the NHS.

However, this is easier said than done. If developing a mission statement and putting written policies and procedures in place were all that was required, it would be relatively simple to change a culture. However, in reality, the espoused values and beliefs of an organisation are often very different from the beliefs in use (‘mental models’) that actually guide behaviour. Mental models are the unconscious assumptions, beliefs and images that guide the way we both perceive the world and act in response to that perception.

Yet, because these mental models are unconscious, we are often unaware of how profoundly they affect

our decision-making and actions. In the 1989 New England Journal of Medicine paper quoted above, Berwick described Foreman 1 as subscribing to the “theory of bad apples” in which quality is improved by inspection, finding the ‘bad apples’ and removing them. Foreman 2, on the other hand, creates the environment in which the people who do the work are given the space to improve the way the work is designed and executed in order to improve its quality.

These are two fundamentally different mental models. In the ‘bad apples’ world, unhelpful mental models are created. Managers may view the people that do the work as not to be trusted and vice versa.

When problems occur, the search begins for who is to blame. Perceptions become polarised and every action is examined for the hidden agenda or motive. In the absence of mutual trust, management

often only exists to ‘command and control’, which is in itself an illusion. Such mental models are corrosive

and often undermine the intrinsic motivation of people working at the frontline. Moreover, when such mental models become entrenched they are very difficult to change.

Foreman 2 has a different set of mental models: people are generally trying their best to do a good job, often in difficult circumstances, and are keen to improve. If leaders and managers consistently base their actions on mental models such as these, they will slowly create an environment in which they are viewed differently by the people they manage, which will in turn lead to changes in behaviour.

Actions speak far louder than words and it is generally easier for people to act themselves into a new way of thinking than to think their way into a new way of acting.

We are what we repeatedly do. Excellence, then, is not an act, but a habit.

Aristotle

Our habits are generally based on our unconscious mental models, and – at the risk of stating the obvious – in order for quality improvement to become continuous, it must be habitual. Habits are triggered by specific cues. In a CQI mindset, a problem becomes a cue to trigger the search for its root causes in order to identify how to permanently resolve it.

Therefore the first half of creating a CQI culture is for leaders and managers to have what W Edwards Deming termed “constancy of purpose” to improve the systems that provide the service. This requires both developing clarity of purpose and vision, and ensuring the mental models that drive leadership behaviour support them. Deming recognised in the 1940s that this meant valuing and developing frontline staff and “driving out fear”. Only when fear and blame is removed can a culture of openness flourish, creating the

environment in which teams can work collaboratively to improve.

In God we trust, everybody else must bring data.

W Edwards Deming

The other half of creating a CQI culture is developing a rigorous data-driven approach to improving the service for patients using the scientific method (Plan, Do, Study, Act – PDSA; also termed Plan, Do, Check, Act – PDCA).

In essence, this means studying the current systems and processes and their outputs; determining the root

causes of gaps between the actual and desired performance; building and testing hypotheses that address the root causes, and acting on the basis of evidence to put in place new processes and monitor the outputs to ensure that performance is both improved and sustained.

It is important to accept that not all PDSA cycles will be ‘successful’, the objective is to test hypotheses in order to find new ways of delivering better service performance.

Each PDSA cycle is an opportunity to learn, including those that fail to deliver improvements in performance. CQI becomes a never-ending series of PDSA experiments seeking perfect performance.

For this to happen, organisations need to create simple and easy-to-use tools that can facilitate team learning through rapid PDSA cycles. Path Links, the service I work within, uses ‘A3 thinking’: a simple, single-sheet proforma for capturing PDSA cycles.

A3 thinking is a powerful, yet simple tool developed by Toyota in the early 1960s to facilitate PDSA. Used properly, it becomes a common currency for systematic, data-driven PDSA problem solving that can be

shared easily with everyone involved in a service improvement project, encouraging collaborative problem solving and learning.

A number of other important and useful tools are used as part of the data-driven PDSA approach. These include process mapping, control charts, Pareto analysis, fishbone diagrams and other root-cause analysis

tools (such as 5 Whys). These tools are relatively simple to learn, but relatively few healthcare staff have been trained in their use.

It is important to emphasise that CQI and the tools associated with it are not solely within the domain of quality ‘experts’. CQI should become part of the routine work of the frontline staff delivering healthcare, supported by quality improvement experts. Each PDSA cycle then becomes an opportunity to develop the CQI capability of frontline staff increasing the capacity to deliver further service improvement.

This is achievable over time if there is “constancy of purpose”. It is not a quick fix. When the Berwick report into patient safety was published in 2013, Donald Berwick wrote an open letter to all NHS staff urging them to develop a new approach to quality improvement. I will finish with the final two paragraphs of that letter.

“We are recommending four guiding principles, among others, to help the English NHS get better faster, and I urge you to think about these and ask how you can help incorporate them into your own daily work.

Dr David Clark

Consultant Histopathologist