As a medical writer, one of the most essential tools to have at your disposal is not the ability to build an argument up from its raw elements to a brilliant conclusion or even to a wide-ranging vocabulary that can make a piece of medical writing shine, but to be able to understand what you are looking at.
A conclusion in a manuscript has no value without evidence, and that evidence has no value if it is not reliable. Consequently, one of the most important tasks, whether for a clinician deciding on a treatment strategy, a researcher gathering data or a provider of medical writing services, is to be able to critically appraise the literature.
Over the years, there have been several attempts to help grade medical literature, and there are a wealth of classification systems and guidelines to aid the busy medical writer. However, one of the simplest and clearest systems is one adapted from the Oxford Centre for Evidence-Based Medicine system. 
This system describes five levels of evidence, based on the design of the study in question, and the definition of each level depends on whether the study was a therapeutic, prognostic, diagnostic, or economic or decision analysis. As a rule of thumb for the medical writer, Level 1 is the highest level of evidence, and Level 5 the lowest.
Level 1 – Randomised controlled trial (RCT) and meta-analyses of RCTs
In an RCT, patients are randomly allocated to different treatment options, whether active versus placebo or one of several active treatments. A meta-analysis is a systematic review that combines the results across two or more studies, and is considered the ultimate source of reliable evidence by clinicians and medical writers alike. However, the quality of a meta-analysis relies on the quality of the RCTs that were used.
Level 2 – Cohort studies
The patients here are grouped based on shared characteristics into ‘cohorts’ and then two or more groups are compared over a period of time. The patients are not randomised and the study can be retrospective or prospective, the latter being preferred as it provides the better evidence, but the former is more easily carried out. They can be susceptible to confounding factors and selection bias.
Level 3 – Case control studies
For these studies, patients from a particular group, such as post-menopausal women, with a specific outcome, such as breast cancer, are compared with patients sampled from the same source population. An analysis of previous exposures can suggest associations with the outcome. These types of studies can be useful in assessing outcomes that are rare or slow to develop, but are susceptible to confounding factors and, for example, recall bias.
Level 4 – Case series
A case series, like a single case report, takes patients who have a particular outcome and reports on what was observed and which variables may be associated with that outcome. There is no comparison with a control group. While it is advisable not to draw conclusions of causality from case series and case reports, they have value in, for example, identifying adverse events and in prompting clinical questions and hypotheses.
Level 5 – Expert opinion
The final level of evidence quality that the medical writer needs to consider is that of the expert opinion. When all the other levels of study quality are available, doesn’t it seem anachronistic ever to use expert opinions? After all, these are based primarily on clinician experience and expertise, which could be seen as little more than anecdotal evidence in the worst examples. Yet, the medical writer working on an assignment should remember that, at their best, expert opinions combine that experience and expertise with the best available evidence. Furthermore, what is a guideline if not the combination of the best evidence with the judgement and opinion of highly qualified experts?
For anyone working on a medical writing project, being able to critically appraise medical literature requires a thorough working knowledge of the levels of evidence, study design and personal expertise. Through those, the medical writer can make decisions on what evidence to rely on when building their arguments.
Get in contact with us to find out more about how we approach manuscript development, and how we use only the most appropriate and reliable literature for our projects. We can therefore develop a range of pieces that can support your arguments and help guarantee success with submissions to editors and publishers.
- Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am 2003;85-A:1-3.