Understanding the rollercoaster of contextual influences on the spread of practices

Photo: http://s3.amazonaws.com/estock/fspid10/77/18/32/statefair-arizona-kevindooley-771832-o.jpg

Within our Implementation Science Group at PenCLAHRC, we’ve been exploring how evidence-based practices implemented in one hospital can be spread to other departments or hospitals in the South West of England. In theory, this sounds simple. Why wouldn’t other places make improvements to care found to work well elsewhere? In actual practice, as quite often can be the case, the experience was not quite so straightforward. So, we sought to study in real-time how context can critically influence the spread of practices – helping in some places and hindering in others.

Working with the South West Academic Health Science Network (SW AHSN), we studied, using qualitative methods, two collaborative projects seeking to spread improvements to healthcare in acute settings: the emergency treatment of acute ischaemic stroke (also involving the South West Cardiovascular Strategic Clinical Network) and the implementation of Patient-Initiated Clinics. We wanted to understand how the differences in context within six hospitals (for acute ischaemic stroke) and three departments in one hospital (for Patient-Initiated Clinics) influenced progress in the spread of these particular practices in healthcare.

To help us explore the influence of context we were informed by one of the many frameworks available in the field of implementation science: the Consolidated Framework for Implementation Science (Damschroder et al 2009). This offered us a taxonomy to develop insights into the factors present and influencing, positively or negatively, the spread of these evidence-based practices. Three of the domains were particularly useful for considering contextual factors: the outer setting, inner setting and the characteristics of individuals.

For the spread of stroke treatment improvements and the Patient-Initiated Clinics intervention, our analysis highlighted the following important contextual influences.

At the macro-level of context (e.g. focus on patients, national and regional influences) we found these influences playing a role:

  • Whether there was competitive pressure to improve as other organisations/teams had already done so.
  • Whether patients’ needs were known, and prioritised, and if implementing the new practices was viewed as beneficially meeting their needs.
  • An absence of national and regional incentives and policies to drive/support uptake.

We found the most influential factors occurred at the meso-level context (e.g. organisation, department and team) and included:

  • Whether the change to practice was a viewed as a priority or strongly needed by each organisation/department.
  • How ready for change was each organisation or department. Particularly if there were skilled and engaged leaders, a stable team, and the available resources (i.e. money, space, and time).
  • Whether, in each setting, could attract, involve and engage key individuals and locate a champion to act as a catalyst/driver for spreading the practice locally.
  • An ability to be flexible (i.e. about timescales and approach) and persist in response to barriers encountered within each organisation/department.

And we found these influences at the micro-level of context (e.g. key individuals within a setting):

  • What key individuals know of and believe about an improvement and intervention.
  • The perceived value key individuals placed on making a change to current practice.

From our work, we generated lessons in the form of questions. These may help to identify contextual factors that could influence future efforts to spread evidence-based practices. We note a need to consider how differing contextual factors and levels interact with each other.

We also identified other factors that influenced uptake. How key people viewed the strength of evidence underpinning the change, and the benefit of external support from university-researchers and, for the stroke project, a quality improvement manager. We identified some challenges too. For example, how to ensure the sustainability of improvements and interventions over time so they become routine healthcare practice.

One of our main lessons was for those involved in spreading these practices to be prepared for hard work and expect the unexpected. So, it’s a bit like being on a rollercoaster! This makes sense when we consider each healthcare setting will have their own demands and complexities that impact on the uptake of improvements and interventions. We will be working with the SW AHSN on how we can build on these lessons to influence future efforts to spread practice. To read more about our study, here’s our project page.


Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implementation Science. 2009;4:50. doi:10.1186/1748-5908-4-50

Pfadenhauer L, Rohwer A, Burns J, Booth A, Lysdahl KB, Hofmann B, Gerhardus A, Mozygemba K, Tummers M, Wahlster P, Rehfuess E. Guidance for the Assessment of Context and Implementation in Health Technology Assessments (HTA) and Systematic Reviews of Complex Interventions: The Context and Implementation of Complex Interventions (CICI) Framework [Online]. 2016. http://www.integrate-hta.eu/downloads

The Health Foundation. Perspectives on context: a selection of essays considering the role of context in successful quality improvement. London: The Health Foundation. 2014. health.org.uk/sites/health/files/PerspectivesOnContext_fullversion.pdf

Is this thing on? Cultivating K-star communication

To mark the 2 year anniversary since the last post (but who’s counting?), we are resurrecting this LKD blog, much to the delight of our one and only commenter (Hi Mark!).

Okay, enough self-deprecation. I am genuinely excited about the new content that will be coming to this blog in the near future, courtesy of PenCLAHRC’s very own Implementation Science Posse™.*

Our group has certainly grown in the last couple of years, bringing in a diversity of experiences, expertise, interests, and perspectives about all things K-star and implementation science. La posse de mosaïque,** if you will.

One platform to facilitate exchange of these diverse perspectives that we have been trialling is a K-star reading group – a club of sorts. It’s just like fight club.

Here are the rules:

The first rule of Fight Reading Club is: You do not talk about Fight Club talk about it with anyone who will listen.

The second rule of Fight Reading Club is: You do not talk about Fight Club.try to convince people to listen.

Third rule of Fight Reading Club: Someone yells stop, goes limp, taps out, the fight is over. sorry, but you continue until the allotted time runs out. Having a coffee before the meeting can help avoid this.

Fourth rule: Only two guys to a fight The more the merrier.

Our first reading group meetings earlier in the year have spurred interesting conversations within our team. As someone who has had to get re-immersed in the literature after some time away, creating the time to read, critically reflect, and debate recent developments in the field with colleagues has been very helpful. I find these discussions are important for me to situate my own perspectives and where I stand in related to current discourses.

In the spirit of further broadening these conversations and fostering the relationships we have with researchers at the University of Western Ontario in Canada, we held a joint reading group with Dr. Anita Kothari, Dr. Shannon Sibbald and their students.

Commemorating our UK-Canada ties with maple leaf shaped burgers.

The discussion proved to be an interesting reflection of the recently redeveloped i-PARIHS framework. Overall, the discussion raised a number of questions and issues, not just of this iteration of PARIHS, but also of the overall state of play of frameworks, theories, and models in IS. With regard to i-PARIHS, there were questions as to whether any substantial contributions were made. Some expected more in terms of providing guidance for operationalization, although it was also pointed out that this was perhaps the theoretical piece of the work, with the operationalization paper to come in the future. Still, others questioned the increased focus on facilitation. An issue raised was that although facilitation is undoubtedly an important part of implementation and its emphasis in the i-PARIHS was an overall positive thing, was the discussion about facilitation left too thin to have a meaningful impact? It was pointed out that much discussion about facilitation reflected concepts in the knowledge brokering literature that were perhaps referred to in this paper with slightly different terms. What was being missed by not referring to and acknowledging the longer history of these discourses in that body of work? There were also a number of questions raised as to other characteristics of facilitators that were not mentioned in the paper. For instance, the facilitators’ position and power to create change, as well as their relationship with the organization (e.g., internal or external) plays an important role in implementation; different characteristics are needed depending on the scale of change that needs to happen. In all, our reading group on the i-PARIHS highlighted some initial points of deliberation and reflection, in line with the authors’request for feedback from the IS community.

An interesting question that was asked – and remains unanswered – relates to why certain implementation science frameworks seem to be more popular in one country versus another. For instance, our Canadian colleagues pointed out that the PARIHS framework is frequently used and mentioned in many Canadian studies, which is certainly not the case in England. One possible explanation offered is the relative accessibility of PARIHS for those who may be new or not well-versed in the field. For instance, in comparing PARIHS with another implementation science model like CFIR, it does not take long to realize that both the language used and the simplicity of the constructs in PARIHS is much more accessible. Another possibility is that one of the critiques of PARIHS as being applied in diverse ways is also a strength: its flexibility allowed for people to use it in a way that was in line with their understanding of the framework, whatever those may be.

Nevertheless, the question remained as to what gives rise to the ubiquity of one theory/model/framework over another in different settings. It was great to be able to have these international(!) conversations with our colleagues. I see these shared reading group discussions as opportunities be cognizant of the trends and developments in the field in these different countries. How do we share and learn from each other? How do we support each other’s work as we advocate for the continued investment in implementation science research in both countries? Alas, these are the big questions our humble reading group will take the baby steps to address.

*No one gave me permission to call us this. Fun fact: We very nearly called our team “Knowledge for Change!” (KFC), but decided against it due to the whole WWF vs. WWF precedent.

**This is probably grammatically incorrect. I’ve mostly forgotten my Canadian French classes.