Objectives It is usually taken for granted that ‘the best available evidence’ is represented by the findings of rigorous scientific research which, in turn, directly inform the development of recommendations for practice in the form of clinical guidelines. We challenge this assumption and examine the role played in the enactment of evidence-based healthcare by other forms of codified knowledge, i.e. knowledge that is formal, systematic and expressible in language or numbers, making it easy to store, transfer and utilise across space.

The study addresses the following research questions:

1. What forms of codified knowledge are seen as credible evidence by practitioners?

2. What are the relationships between these forms of knowledge in the enactment of evidence–based practice within healthcare organisations?

3. What is the impact of these forms of knowledge on evidence–based practice?

4. How do the composition and impact of codified knowledge vary across different high–income countries?

Method This exploratory study emerged from a broader research programme examining leadership and facilitation in the implementation of evidence-based nursing across the UK, Australia, Canada and Sweden. Within each country, up to two organisations were selected based on the following criteria:

1. self-declared adherence to the implementation of evidence-based nursing;

2. adequate organisational performance; and

3. broad access to several levels within the organisational hierarchy granted to the researchers.

55 research participants were recruited to represent different levels of the hierarchy, roles and sectors. Semi-structured interviews served as the main method of data collection. Data analysis was organised in two stages. The first stage, focusing on the construction of country-specific narratives, combined the codes derived from the interview guide with descriptive codes that emerged inductively. The second stage utilised the deductive coding framework informed by the literature and applied across all four datasets. Matrix analysis was deployed to facilitate cross-case analysis.

Results We argue that research evidence and its direct derivatives, such as clinical guidelines, are NOT the dominant forms of codified knowledge deployed in the organisational enactment of evidence-based healthcare.

We describe the chain of codified knowledge which reflects the institutionalisation of evidence-based healthcare as organisational ‘business as usual’. This chain is dominated by performance standards, policies and procedures, and locally collected (improvement and audit) data, i.e. various forms of ‘evidence by proxy’ which are, at best, informed by research partly or indirectly but are nevertheless perceived as credible evidence.

Our cross-country analysis highlights the influence of macro-level ideological, historical and technological factors on the composition and circulation of codified knowledge. Prioritisation of ‘evidence’ by proxy’ and marginalisation of clinical guidelines are likely to be more prominent in those countries, whose healthcare sectors have historically been more engaged with the New Public Management logics of standardisation and performance measurement.

Conclusions Our analysis reveals dual effects of this codification dynamic on evidence-based healthcare. On the one hand, the legitimisation and mobilisation of contextual and local knowledge counterbalance ‘dogmatic authoritarianism’ apparent in the more restrictive interpretations of ‘evidence’ and potentially enable bottom-up knowledge flows. On the other hand, this is achieved through a significant dilution of the initial paradigm, excessive formalisation, and detachment of frontline staff from the fundamental competencies and knowledge base of evidence-based decision-making, whereby direct use of research evidence and clinical guidelines is becoming a prerogative of experts, represented by professional elites and designated facilitators.