Each of these reviews results in reports of varying depth and structure but they do have some common principles. They are analyses of the currently available evidence organised in narrative form alone or with the addition of tables¹. They highlight gaps in the evidence base and opportunities for further research. More comprehensive/exhaustive reviews assess literature in accordance with pre-defined inclusion/exclusion criteria.

Literature reviews are indispensable to researchers in the healthcare and pharmaceutical industries. They are often conducted at the beginning of every evidence-generation activity. Reviews can be undertaken across different areas, from assessing healthcare pathways to evaluating support for a medical device or drug. Narrative and scoping reviews can help to form the basis for further research, by highlighting current problems and a way forward. Systematic literature reviews and meta-analyses are more in-depth and can provide a rigorous view of current research and its limitations. In turn, each type of review can be used to create a platform which can inform policy making and drive change.

In this blog, we’ll look at the 5 common mistakes to avoid when conducting a literature review.

1. The research question is too broad

Undertaking a successful, publishable, literature review requires a focused research question. It is important to think of the target audience and the problem that is to be addressed. Key aspects to consider in a research question are²:

Using the “PICO” format is useful when constructing a research question in healthcare³. This helps by identifying the: Patient problem/Population, Intervention, Comparison, and Outcome. For example:

In cardiorenal patients who develop hyperkalaemia, should RAASi therapy be maintained or discontinued in order to successfully reduce blood potassium levels?

2. Unclear inclusion/exclusion criteria

Selection criteria for literature reviews are a set of inclusion and exclusion guidelines for the types of articles which will be examined. They can depend on the style of the review being undertaken. Whilst narrative reviews may require a historical understanding of a topic and include seminal research, a systematic review may be more targeted. A common inclusion criterion is that studies must have been published within a certain time frame (e.g., the last five years).

Selection criteria should be based on the research question. If the target research population is based in a specific country, it may be necessary to exclude international research. The inclusion guidelines should state elements of a publication which must be present to be included, with exclusion guidelines stating which criteria will disqualify a study from inclusion⁴. Some examples are listed below:

3. Choosing the wrong sources

Narrowing down literature for inclusion in more systematic reviews involves two stages. Initially, any duplicated results will be removed, before each study is individually analysed to ensure it meets the inclusion criteria. When analysing studies against predefined inclusion and exclusion guidelines it is important to assess the quality of the study⁵. It can be helpful to create a quality assessment checklist before any searches are undertaken. Quality assessment will provide a basis for further analysis and help when constructing the discussion for the final publication. Quality assessment may also provide a basis for additional inclusion and exclusion criteria. However, scoping reviews or mapping reviews with time constraints may not have explicit quality guidelines. Examples of aspects of research quality are illustrated below:

The critical appraisal tool coined by the Centre for Evidence-Based Medicine is designed to help researchers systematically assess the quality of clinical evidence based on their research question⁶. The tool asks four questions of clinical evidence:

If the answer to any of these questions is no, then the evidence is not suitable to include in a review.

4. Not leaving enough time to conduct

Literature reviews can be time-consuming to undertake. As such, plenty of time should be put aside for each stage of the research. If a literature search is rushed, then researchers will risk missing salient studies to include. Conversely, it could also mean that literature which is not relevant or of a high standard will be included in the review. More time should potentially be allotted for planning and analysing, as careful consideration at these stages will provide a good foundation for writing up.

The stages you should consider are illustrated below:

The time needed to search for literature should not be underestimated. The analysis of references within publications may lead to further research for inclusion, and therefore greater search time⁷.

5. Inappropriate synthesis of the research findings

When writing up a literature review always bear in mind the target audience and the problem at the heart of the research⁸. Reports should always reflect the nature of the data gathered, and any conclusions must be aligned with the results. Overstepping the evidence and creating a false narrative from the data can have serious consequences. It is important to assess the weight of the evidence gathered and be able to discuss the limitations of included publications. However, synthesising results is more than critiquing each research paper.

Sources selected for review should be organised according to their themes, e.g., similar interventions, outcomes, and difficulties. Using a review matrix or table helps to localise related information for assessment⁹. Themes should be discussed systematically, connecting ideas, and showing how different elements influence one another. This will enable the target audience to see the combined scope of the assessed research and problem area. Finally, the results of the review should be contextualised with a comparison to other relevant studies. A simplified work process for synthesising data in a literature review is shown below:

Key takeaways

Overall, when conducting a literature review it is important to make sure that:

A strong literature review will inform the nature and scope of further research so that evidence generation can be approached strategically to see better decisions actioned in healthcare.