By the mid-19th century it had been convincingly demonstrated that handwashing by healthcare staff dramatically reduced patient infections, but only in the 20th century was it mainstream practice. Even in the 21st century best practice is far from universal, as NICE lamented in its 2014 Quality Standard on the subject.

Of course not everything takes two centuries to spread. On the one hand, a study of healthcare interventions from the 1960s to the 1990s found it took an average of 17 years until they became standard care.(1)

But on the other, some practices go ‘viral’ very quickly. For example, featured in The Health Foundation’s recent Power of People film(2) was the Flo’ text messaging app, which enables patients to receive reminders and health tips via their mobile phone. Through a social enterprise – Simple Telehealth – which licenses the package, it has rapidly spread to more than 70 organisations and continues to grow. Another innovation that the Health Foundation has supported, PROMPT – which is a one-day, team-training package to reduce maternal and child morbidity around birth – has within a few years spread across the UK and well beyond.(3)

There are myriad reasons why proven innovations and improvements can spread at different rates and why demand for them can differ – ranging from their complexity and the cost and effort of implementation, to the nature of their evidence base and the persuasiveness of their advocates. But underpinning many of these issues are the complications of context.

Adoption versus adaption

A basic issue here is conceptual. It is usual to consider innovations as discrete entities that simply need to be adopted elsewhere to have impact. This is akin to thinking of an innovation like a pill that only needs to be taken to have its effect. Under this thinking, the main issue to facilitate spread is to motivate staff to take the ‘pill’ – i.e. to copy and implement the exact same innovation – and to consider which factors best increase that motivation. For simplicity, let’s call this a copy-and-paste ‘adoption’ view of the world.

But to continue the analogy, we now know that the impact of a pill can differ according to not just the age, sex and weight of an individual but also their genome. In other words, ‘context’ – the environment in which the innovation or intervention is meant to operate – matters. In healthcare many factors can make the environment helpful or hostile to the implementation of new ideas or interventions.(4) And what is more, to be successful an intervention may need to be adapted to the context in which it is implemented, or the context itself can change the intervention. Let’s call this the ‘adaption’ view of the world.

The adoption view is relatively simple. It may work for some families of innovations, for example the simple, possibly intuitive, often technology-based and which don’t rely on too many people to change behaviour or too much extra investment beyond the cost of the thing itself. For example, a new bit of diagnostic kit, or some kinds of app. Those with a simple adoption orientation may favour simpler motivators – external ‘push’ levers and nudges to motivate healthcare staff or patients to do what seems absolutely obvious, such as more information, mandated guidance, targets, financial incentives and regulation. This worldview also resonates with the drive to ‘standardise’ key aspects of care.

The adaption view of the world, by contrast, is more complicated. Some families of innovations are complex, may not be discrete, must be adapted to the local context and rely on different staff groups and patients to develop them. For example, innovations to improve the flow or safety of patients along a clinical pathway.(5), (6) As we see with many of the projects the Health Foundation funds, this work is often a long, hard grind, requiring intensity of local effort that does not tail off after a few months. Because of this, those with an adaptation orientation may favour ongoing ‘pull’ factors internal to the provider to clear the obstacles for staff and patients to maintain energy and focus constantly to design and deliver. These factors may be many, more diffuse and more local, such as leadership and trust, which allows autonomy and risk-taking to adapt interventions over time; a collaborative working culture if many staff groups are involved in designing and making changes; and rapid feedback and learning loops relying on continuous assessment along with the space for intelligent course correction. The ‘adaptors’ will argue that spreading best practice is much more complex than a simple concept of ‘standardisation’.

Of course these are overly simple twin archetypes, roughly caricatured as authoritarians versus free thinkers, pragmatists versus idealists, rules versus trust or mechanistic versus organic modes of organisational behaviour. The truth is that both approaches are probably needed in most cases, but in proportions we don’t yet know enough about. At present we are shooting in the dark when trying to spread innovation because we are still treating it as a pill to be taken everywhere.

What is now needed is a more intelligent typology of innovation, host context and motivators – and even more intelligent analysis as to their interaction and impact. Only then will we be able to chart a speedier course to spreading innovation and improvement.

Where do we go from here?

All this sounds complicated, and to the pragmatic (who just want a handbook, and now) possibly indulgent; after all, the quick fix is always the sharp nudge. In the immediate term, some important insights from David Albury at the Innovation Unit might help.(7)

First, supported interacting communities of innovators and potential adopters/adapters can help develop and spread innovation and good practice. Such communities are structured, facilitated and supported to use disciplined methods to develop, codify and adapt innovations. This is exactly the kind of approach we will be looking at in our Improvement Labs, which will bring people together to make progress on complex health challenges – part of the ‘Q’ initiative that the Health Foundation is developing with NHS England.

Second, mobilising ‘demand pull’ through movement building. This requires building a compelling case for change alongside a vision that inspires people, as well as developing and empowering wide coalitions to call for change.

Third, creating an ‘enabling ecosystem’ for spread, including a culture and leadership that encourages experimentation; investment and infrastructure to support spread; incentives and rewards for adoption and adaption; and an enabling policy environment that allows local freedoms.

Try all this in an environment of austerity, when ‘grip’ is in the ascendancy? But that is the task now at hand, for the goal of £22 billion efficiency savings will remain no more than wishful thinking if innovations and best practice stay locked in their site of origin. And ‘first base’ must be to understand better the nature of the spread challenge.

References

1. Balas, E.A. and Boren, S.A. (2000) ‘Managing clinical knowledge for health care improvement.’ In Bemmel, J. and McCray, A.T. (Eds.) (2000) ’Yearbook of Medical Informatics 2000: Patient-Centered Systems.’

2. www.health.org.uk/flo

3. www.health.org.uk/programmes/improvement-science-fellowships/projects/tim-draycott-improvement-sciencefellow

4. www.health.org.uk/publications/perspectives-on-context

5. www.health.org.uk/publication/improving-patient-flow

6. www.health.org.uk/programmes/safer-clinical-systems

7. Albury, D. (2015) ‘Myths and mechanisms: a brief note on findings from research on scaling and diffusion.’ London: The Innovation Unit.