User:LGreg/sandbox/Approaches to Knowledge (LG seminar 2020/21)/Seminar 18/Evidence/Evidence in medicine

Role of evidence in medicine: evidence-based medicine
Evidence constitutes an integral aspect of medicine: Indeed, a key proponent of the discipline's current paradigm is evidence-based medicine (EBM), defined by Sackett et al. as 'the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients.' By using said 'current best evidence', EBM aims to eliminate (if not, minimise) any biases that may have arisen due to conflicts of interest. In medicine, external evidence from research is not the sole operative factor behind clinical decisions, but rather is used alongside individual patients' wishes, in conjunction with the clinical expertise and experience developed by practitioners over time (internal evidence), in order to optimise healthcare outcomes.

How external research evidence is obtained & studied
As asserted by Webb and Masic et al. respectively, the type of external evidence utilised in medicine is empirical (typically obtained from observations and experimentation), and acquired through clinical research. This research follows the scientific method, whereby results from experiments undergo statistical analysis, in order to determine their significance. If results are determined non-significant, it is implied that there was insufficient evidence to make inferences. On the contrary, if the results of a study are deemed to be significant, they are used to either accept or reject a null hypothesis, allowing researchers to draw conclusions : This is regarded the scientific evidence required by clinicians to make decisions regarding treatments.

However, a crucial feature of evidence-based medicine is that, the 'best evidence' is used. Thus, before external evidence can be applied and implemented into practice, the scientific literature (usually retrieved from electronic databases) must be critically evaluated and assessed for its validity and utility or relevance - a process termed critical appraisal. Critical appraisal scrutinises the research methodologies used in each experiment to determine the quality of evidence, by studying the consistency of conclusions made against available results, the repeatability of results, and the selection of statistical analysis used.

Methods to gather external research evidence
There are many different types of research methods and studies used in medicine to gather external evidence (which could be either filtered or unfiltered), each with its own strengths and limitations. Some of the principal ones are described below.

Randomised control trials
Randomised control trials (RCT) are used in medicine to determine the efficacy of new treatments being developed. A method of primary data collection, RCTs are often regarded the 'gold standard' of clinical research, considering they help reduce selection biases by the researchers, given that the sample population is allocated randomly - often using a number generator. The new intervention that is to be tested is given to one group (the experiment group), whilst a second control group in the trial is given either a placebo treatment, or a pre-existing one. This is done double-blindly - neither the researchers nor the participants are aware of which group has the newer intervention - in order to reduce the chance of participants experiencing a placebo effect, or clinicians unintentionally prioritising the experimental group (known as a confounding bias). However, despite such strengths, the use of RCTs to gather evidence for certain sub-disciplines of medicine may be limited. This is the case for surgical interventions, where blinding is not possible. Moreover, blinding in RCTs may also not be achieved if an intervention has distinguishable side effects. These factors limit the extent to which RCTs may be used to gather evidence in medicine.

Systematic review
A systematic literature review focuses on a research question, for which it collates and appraises a range of studies. Results may be presented in either a qualitative or a quantitative approach. Systematic reviews may be used to collect secondary evidence in medicine as they are 'methodical, comprehensive, transparent, and replicable'. Due to this incorporation of multiple sources, which are all critically assessed, with their inclusion in the review justified, systematic reviews tend to present evidence in a more objective and neutral manner. A notable example of the use of systematic reviews in the discipline of medicine is Cochrane, an organisation which publishes such reviews on emerging healthcare interventions.

Meta-analysis
Meta-analyses are statistical means of collating data for a particular research question from a range of different studies (these are secondary sources), to form a weighted average. Whilst meta-analyses may be carried out for any grouping of multiple studies (so long as they have the same research focus), they are often a constituent of systematic reviews, however not necessarily always, as the latter may also be qualitative, whilst meta-analyses are purely quantitative.

Other
Other research-based methods of generating evidence in medicine include: cohort studies, case control studies and case reports and series, all four of which are observational techniques.

Internal evidence
Clinical expertise, or expert opinion, is considered to be a form of qualitative, internal evidence used in medicine to help practitioners develop treatment plans. It is subject to each clinician's own experience in the field, accumulated over the length of their individual career. Unlike external evidence from research, internal evidence is not validated, and hence is more unreliable and fallible.

Levels of evidence
A vital component of EBM is that all evidence is ranked in a hierarchy, also known as levels of evidence, which classifies evidence on its perceived extent of bias. Multiple hierarchies of evidence have been developed by key contributors in medicine.

One approach is the University of Oxford's CEBM's 'Level of Evidences'. Whilst there are variations in rankings depending on the objective of the study, systematic reviews with homogeneity (when studies used have congruous results) are consistently placed at the top (Level 1A), and are thus considered to be the most reliable source of evidence in medicine. On the other hand, unverified clinical expertise consistently ranks at the bottom (Level 5). More detailed information on these levels may found on CEBM's website, mentioned in the further reading section.