The Evidence-Based Hierarchy: Is It Time For Change? A Suggested Alternative

What Is in a Question? 

The first step in EBM is formulating an answerable clinical question based on the perceived knowledge need. These questions fall into 1 of 8 categories as shown in Fig 2.

View Large Image Download to PowerPoint

Fig 2. Clinical question categories based upon works by various authors.19, 40, 41, 42

Because these questions are all different in nature, they require different types of research evidence to best answer them. Thus, the idea of one evidence-based hierarchy to encompass them all has been questioned.4, 5, 9, 10, 11, 12 One of the problems with the current hierarchy model is the exclusion of qualitative data in the evidence-based decision making process. This problem has been recognized before13, 14, 15, 16, 17, 18, 19 and has become more pressing with the recent increase in popularity of qualitative research methods.20 The profile of qualitative data has also been increased by the development of the evidence-based patient choice model of patient care in which the external evidence is assessed by the patient and the practitioner together and the decision making process regarding therapy decisions is shared.21 The evidence-based patient choice model requires that the practitioner understands the patients perspective and experience of their condition; the best type of research evidence for this type of information is gained from qualitative studies.22, 23

The authors believe that the evidence-based hierarchy needs to be changed. This change must reflect that the nature of posed clinical question determines the external evidence most appropriate to answer that clinical question. Any alternative offered needs to be simple to follow so that clinicians could use it as a “ready reckoner” on their desktop. This would remove at least one of the perceived barriers to the implementation of EBM through helping the practitioner find the best research evidence with which to answer their particular clinical question.24, 25 The devised evidence pathways were designed based on the Centre for Evidence-Based Medicine detailed tables of evidence26 and on the qualitative research literature.13, 15, 16, 18, 22, 27, 28, 29, 30, 31 The usefulness of any evidence pathway is limited if used alone because it is vitally important to follow with appraisal the external research literature (Fig 1, step 3) before its use in the decision making process, rather than ranking it solely by any hierarchy. As Greenhalgh32 says, “…not even the most hard line protagonist of EBM would place a sloppy meta-analysis or a RCT that was seriously methodologically flawed above a large, well designed cohort study.”

Systematic Reviews 

The role of the systematic review is well established within the quantitative paradigm as a means of improving credibility and generalizability of research data and distilling these data into a format that is easy to access and understand by the clinician.33, 34, 35, 36, 37 They are considered the highest level of evidence and are often used as the basis for health-care decisions from government level down.35 To date, these reviews have focused predominantly on effectiveness of treatment interventions and so have been limited to RCTs.35 A systematic review of qualitative data has been attempted,38 and a framework for such systematic reviews is under development.13, 14, 15, 16, 35, 39 However, even the development of a qualitative systematic review framework is controversial because of the context sensitive and highly interpretive nature of the data. The evidence pathways presented in this article use the idea of systematic reviews for both the quantitative and qualitative paradigms in the expectation that acceptable frameworks for systematic reviews of the qualitative literature will be formulated and published.

Evidence Pathways 

The evidence pathways presented here stem from the 8 types of original clinical questions listed in Fig 2. For convenience, clinical questions that share research evidence pathways are grouped together; these pathways are shown in Fig 3.

View Large Image Download to PowerPoint

Fig 3. Evidence pathways. Gold is considered the highest level of evidence. Silver is considered good level of evidence. Bronze is considered the weakest level of evidence. The area outlined with box represents the qualitative methodologies.

The clinical question is the focus of the pathways, and the research evidence appropriate to that type of clinical question is ranked according to the strength of evidence they provide. The best evidence available is ranked as gold; the next best evidence is ranked as silver, and the bronze ranking covers the weakest evidence. The ranking of the type of evidence assumes that each is of good quality and that each individual piece of research should be assessed for methodological quality before deciding upon its importance in determining patient care (Fig 1, step 3).

The qualitative section of the pathways (outlined with a box around that section in Fig 3) should be considered in the light of considerable overlap between the research types. For instance, if the clinical question was “What it is the experience of suffering from a migraine?” it is likely that phenomenological grounded theory and ethnographic methodologies could be used to answer the question. This is because, if a researcher asks a group about their migraines, this group of “migraneurs” could be considered a cultural subgroup, and therefore, the results could be published as an ethnography article. Likewise, if a researcher sets out to develop a theory about persons with migraine, the process may well uncover important information regarding the lived experience of a migraine but be published as a grounded theory article. These papers would be missed if the clinician limited their literature search (Fig 1, step 2) to phenomenology papers only. However, because phenomenology papers on this subject set out to discover lived experience, it is more likely that the information gained from phenomenology papers would be more thorough and relevant. Therefore, it is still worth looking for phenomenology papers first when trying to answer a clinical question regarding lived experience.

Conclusion 

These pathways may help to guide the user to the most appropriate research for the type of clinical question being asked. They are not all-inclusive and are kept simple in an effort to make them easy to use. It is hoped that the presentation of these pathways will stimulate further dialog in the health-care professions as to the role of qualitative research methodologies in EBM.