Over the last 20 years, in the world of evidence-based health care, there are running debates about what constitutes evidence and hidden assumptions in the way the word evidence is used. In defining this word, it is important to recognise that there are different types of definition. The simplest descriptive definition of the word evidence is defined by the Oxford dictionary as the available body of facts or information indicating whether a belief or proposition is true or valid.
Facts or information are the starting point of evidence. Yet, information can mean different things to different people, and in different circumstances. It may be that information constitutes evidence in one setting, but not in another.
When used in the phrase, evidence-based, a different definition is offered as denoting disciplines of health care that proceed empirically with regard to the patient and reject more traditional protocols. It seems that by using the whole phrase evidence–based changes the emphasis of evidence to preferentially value scientific information.
Further, our understanding of evidence may depend on what we want to know, why we want to know it and how we will use it. The potential use of evidence is well explained in an early definition … Evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. Implicit in this definition is the recognition that doctors require additional information from and about patients and they also need to understand how to make judgements about what is the best scientific evidence. Could this be a key to the debate? Can information about patients’ needs and clinicians’ expertise be defined as evidence? While this information may constitute evidence according to our first definition, it may not concur as well with the definition of evidence-based.
However, information about patient conditions and preferences, and the clinicians’ expertise is required to interpret and use research evidence. The term practical evidence was used to describe judgements made about the meaning and applicability of research data in the context of information from personal experience, advice from colleagues and local norms of practice.
This argument has been well developed, such that there is now widespread opinion that research evidence is not sufficient to inform clinical practice or healthcare policy. Ten years ago a key nursing paper argued that there are 4 types of evidence; namely research, clinical experience, patient experience and information from the local context, and that information from all 4 sources are necessary for making health care decisions. Although widely cited, this paper has had limited influence. It seems that the EBM movement continues to prioritise information about research, calling it evidence while implicitly downgrading information about patients and clinicians. Furthermore, information about the actual context of clinical practice has been assigned to a black hole, otherwise known as the knowledge-practice gap.
Can we describe evidence sufficiently by definition alone without understanding how it is being used? Perhaps it is useful to recognise that the word evidence has adopted an implicit acknowledgement of scientific rigour, when it is used in the phrase evidence-based. However the way in which evidence is being used is important, and the elite dominance of research evidence in clinical decision making needs to be contested.
It is important to realise the limitations of research evidence, such that average study results do not always apply to target populations. Additional knowledge is required to apply research results in particular cases. This knowledge requires key information about patients, clinicians and the context in which they work. There is a need to collate and organise practical, experiential and theoretical knowledge to judiciously use the research evidence to make the best decisions for patients.
There is a final challenge to this argument – when empirical research underpins information about patients, clinicians’ experience and their local contexts, can we call this information evidence, especially if we can demonstrate how it influences the use of research results? Should we recognise a pragmatic working definition of evidence for improving health care?
I conduct and supervise research that promotes the translation of quality research evidence for use in clinical practice.
KT@OX 
thanks Anne, a helpful comment and reference, to really begin to understand the health interventions that we read about.
Whatever we call ” information about patients, clinicians’ experience and their local contexts” without it we cannot understand the issues arround commissioning/decommissioning health services and certainly cannot lift and shift “best practice” because these are the variables which determine whether or not a service will have the desired outcomes. At the moment we have no systematic way of appraising those aspects of health services because they are not been documented adequately when writing up case studies.
The new NIHR guidance to researchers “Developing a checklist for research proposals to help describe health service interventions in UK research programmes: a mixed methods study” http://www.health-policy-systems.com/content/12/1/12 may provide a useful starting point for not only when setting up the research but also in the write up and may help us to fill in the gaps in our understanding of existing service case studies.