In the Getting started with research, I discussed means of getting started in research, choosing a research area and development of a research question. This leads the research student on to the development of a research proposal, however, before the final research proposal can be developed and presented, it is imperative that the previous literature in the area is scoped and evaluated to identify the need for the proposed study and establish its context in the specific research area. In this article, I will discuss means of undertaking and presenting a literature review. While the focus of the article is presenting the literature review for dissertation purposes, I will also touch upon disseminating review in academic journals.
What is a literature review?
A literature review has been defined as a process of searching, reading, analysing, evaluating, summarising and presenting the body of evidence related to any specific topic in a sequential manner. It should be considered as a means of undertaking a thorough, objective and critical examination of currently available research literature on a topic. It should not be considered as an act off simply finding a few articles and writing up their results in a few paragraphs to form the literature review of your dissertation/ thesis. On the contrary, a well written literature review tells a complete story about any topic in a succinct, clear manner and presents a clear argument for the need for your research by highlighting the gaps in knowledge.
For e.g. Sourav wants to undertake a study exploring the effectiveness of a targeted exercise program for low back pain, it would not be sufficient to look at some articles which have used a similar form of exercise program, but the literature review would also need to look at current knowledge in relation to pathomechanics of low back pain, current exercise approaches for managing low back pain, mechanism of action of exercises in low back pain and finally, present any studies which have explored the same technique and lead the readers of dissertation to why his study is needed.
Why undertake literature review?
A literature review should not be seen as a chore or a formality to be completed when writing up your dissertation/ thesis. It is an essential task which should be aimed at fulfilling several aims –
- a) Before you embark on your research journey, a literature review provides baseline guidance about what is known about the area of your research. It is important to identify what research has already been conducted and what is known, so that you are not ‘re-inventing the wheel’.
- b) A logical, well presented argument in a literature review shows that you understand the area of your research well. For e.g. while Sourav plans to undertake study on low back pain, if he does not understand how prevalent low back pain is and its impact on an individual or society, it is unlikely that he will be able to justify the need for his research. Therefore the literature review highlights what the most important issues are and their relevance to your work, your knowledge of the controversies in the area, and have the anticipation of where it’s being taken. All this would allow you to map the field and position your research within the context.
- c) A literature review is not only about what is known in the area of research, but also about the unknown, the unrealised or the misinterpreted. This should not be construed to be daunting, but it is infact an opportunity for students. As you critically examine the pieces of evidence, you will realise the unknown, unrealised or misinterpreted, thus leading you to evaluate the need for your own research question.
d)A literature review also allows you to establish your own theoretical framework and methodological focus. As you continue to explore previous research, you will come across studies from different research paradigms, philosophical positions and using different research methods. This will help you to examine your own research focus and the proposed methods of data collection and analysis.
- e) Some undergraduate and graduate (master’s) level courses allow you to undertake a literature review as your research method. It can be quite useful as it does not require approval from local or national ethics committees, which in itself can be a time consuming and laborious process. It is useful to clarify with your research supervisor before making a decision to use literature review as a method for your dissertation.
- f) Finally, a clinician might decide to undertake a literature review to update their current personal knowledge on any topic.
Types of reviews
While most students utilize narrative reviews as a means of presenting the results of their literature review, it is important for students to understand what different methods of reviews are available in order to select the best fit for their needs. Grant and Booth (2009) have presented a summary and critique of 14 different types of reviews which have been utilized within the domain of healthcare research. These include –
- Critical review – The aim of a critical review is to extend beyond simple summarizing of included studies to critically evaluate the quality of included literature. Usually not systematic.
- Literature review – A broad review of published literature which is typically narrative and may or may not include quality assessment.
- Mapping review – Graphical or tabular review which aims of map out and categorize existing literature and is used to identify the need for further reviews or new studies.
- Meta-analysis – A method involving statistical pooling of the results from studies including comprehensive search and quality assessment.
- Mixed methods review – A review which combines results of quantitative and qualitative studies.
- Overview – It is a generic term to explain the characteristics of included studies in a broad review.
- Qualitative evidence synthesis – It is a review which combines and reviews results from qualitative studies.
- Rapid review – It is a time limited review of existing literature, but utilises a systematic approach.
- Scoping review – This is a preliminary review to ‘scope’ out the nature of research available in a given area.
- State of the art review – These reviews tend to address current matters and seek to offer new perspectives on any issue or point.
- Systematic review – It involves systematic methods of searching for, appraising and synthesising research evidence.
- Systematic search and review – It is a combined form of critical review and systematic review to develop ‘best evidence synthesis’.
- Systematized review – These are usually conducted as PhD projects which include elements of systematic review, but cannot be classed as such.
- Umbrella review – refers to review compiling evidence from multiple reviews into one accessible and usable document. Focuses on broad condition or problem for which there are competing interventions and highlights reviews that address these interventions and their results
Thus it can be seen that there are various modes of undertaking and presenting a literature review. Generally with undergraduate projects, time allotted for research will be limited and therefore the focus will be on getting the main research started. But with MSc and PhD projects, students need to understand the nuances of each type of these reviews before deciding on what to include in their dissertations. For e.g. several Msc students have undertaken systematic reviews as their dissertations while as a part of my PhD project, I am currently undertaking a mixed methods research synthesis which involves a combination of a systematic review, a meta-ethnography ( a form of qualitative synthesis) and then combining the results from these into a cross study synthesis.
Step by step methods for undertaking a literature review
Though exact methods for undertaking a review will depend on the type of review chosen, as discussed above, I will describe a general method of undertaking a review which will include developing a review question, searching for literature, evaluating quality of literature and synthesis or analysis of literature. We have already discussed means of developing a research question in the previous article of this series. Similar means can be utilized for developing a research question for review or if the review is the initial part of the same project, the same research question can be utilized.
I will use one of my previous projects to demonstrate steps which should be undertaken to complete a literature review. The aim of the literature review was to explore the role of exercises in pregnancy related pelvic pain.
Searching for literature – In the modern times, the easiest way of searching for some literature on a topic is to ‘Google’ it. But when undertaking a literature search, it is imperative that the search is comprehensive (so that we do not miss useful articles on the topic), systematic (so that it can be re-produced to update your review at a later date) and focussed.
Development of electronic databases has made the task easier, thus overcoming the tedious and time-consuming process of manual searching through databases. It also has the advantage of making the search comprehensive rather than limited by the number of subscribed journals. So the first step is to identify the electronic databases which should be searched. Generally, physiotherapy literature can be found through these electronic databases –
- Web of Science
Beyond these electronic databases, scientific search engines like Google Scholar and Scirus also are an important source of literature searching. In addition, Directory of Open Access Journals (DOAJ) should always be searched for open access journals which might not have been listed in any of the traditional electronic databases.
Having decided on the databases, the next step is to identify the key search terms. Development of search terms is based on specific key words related to the topic. As our topic is role of exercises in pregnancy related pelvic girdle pain, the key concept terms which come to mind include “pelvic girdle pain”, “instability”, “lumbopelvic complex”, “sacroiliac joints”, “pregnancy”, “exercises” and “rehabilitation”. In addition to these, thesaurus terms and synonyms for each of these keywords should also be included in the search. These components should be combined using Boolean logic (AND, OR, NOT).
For e.g. (Exercise OR Rehabilitation) AND (‘lumbo-pelvic complex’)
MEDLINE is the most comprehensive database of health related literature and therefore further information about search strategy will be in relation to this database. There are minor variations which need to be considered when searching other databases. You have the option of searching only the titles or abstracts or both. [AB ( therap* AND relationship ) OR TI (therap*)] will search for terms including therapy/ therapist in either abstract (AB) or titles (TI). Essentially a search strategy is a series of such commands for the database search engines to retrieve relevant articles. There are detailed tutorials for development of search strategy which can be utilised to create your own for each database.
Beyond electronic literature searching, it is also imperative that you manually hand search other journals which might not be indexed in such databases e.g. Journal of Indian Association of physiotherapy.
Beyond journal searching, other techniques for retrieval of literature include searching for unpublished theses (this might be important if you are undertaking a PhD project and want to look at all aspects of your topic which have been covered in previous projects but might not have yet been published). Theses repositories like ETHos (United Kingdom) and TROVE (Australia) record and provide access to thousands of theses which can be accessed online (may need subscription).
Reference lists of articles, review papers and textbooks should be scanned for additional articles. Experts in the field and key authors can be contacted for additional references. Their CVs can be searched on internet to ensure lists of publications are complete. Another important method of searching is citation matching of included articles. You should also undertake citation matching on Google Scholar for all the included articles to find any recent articles which have cited the selected articles and thus may be relevant to your topic.
In summary, literature search should be systematic and each step of the search should be recorded to show the list of titles reviewed, reasons for inclusion and exclusion. This will ensure that you are able to justify your selected articles based on eligibility criteria of your review.
Selection of articles is usually a three phase process with initial screening of titles, then screening of included abstracts and finally retrieval and screening of the full text papers.
Data extraction– Once the articles have been selected for inclusion, the process of data extraction begins. The type of data to be extracted usually depends on your research question. For e.g. as I was looking at the effectiveness of exercise in pregnancy related pelvic girdle pain, I intended to present a short summary of the paper in my literature review section about the number and types of patients studied, the intervention provided, the outcomes accessed and the results. I also needed to critique each paper on its merits and shortcomings. So I decided to create a standardized table which included sections related to these categories and was able to extrapolate data from each of the included study. It is a tedious process which needs to be undertaken with great care ensuring right form of data is captured.
Quality assessment – In order for you to be able to make judgment about the results of any study, it is important that you are aware of the quality of study itself. There are various standardized tools available for quality assessment for different study designs. For e.g. in my study, I included only randomized control trials and therefore used PEDro tool for quality assessment. There are similar forms of tools available for non randomized studies, reviews, qualitative studies and mixed methods studies as well. When you are setting the criteria for studies to be included, you should be able to clarify which tool you will be using for quality assessment of the studies. Essentially any quality tool will look at the methodological rigour of the included study, how it was conducted and robustness of the results, thus giving an indication of how much confidence we can have in the findings to be able to utilize them in our own practice or research.
Presentation –Systematic reviews tend to be presented in a sequential fashion describing the background, methods including literature search, selection of studies, data extraction, quality assessment and analysis. Generally, literature reviews are presented as a form of narrative if they are part of course work or degree dissertation and allow some leeway in terms of their structure and flow. In these, it is important to present a brief introduction, setting the tone for the review and describing the aims, then you can discuss the included studies. It is important to utilize critical writing skills instead of simple paraphrasing and description of included studies. In the paragraph below, I present an example of presentation of results of a study included in my review of pregnancy related pelvic girdle pain –
Wedenberg et al (2000) in their study on 60 pregnant women concluded that acupuncture provided a better option over physiotherapy as an intervention for pregnancy related PGP and low back pain but it suffers from lack of methodological rigour . There is a discrepancy between the two treatments (PT- 10 sessions over 2 months in groups, Acupuncture – 10 sessions over 1 month individually). There is no attempt to follow up whether the home exercise program was followed. There is no comparison with ‘no treatment’. The data presented is insufficient to draw conclusions.
While I describe the results of the study, I also provide my own opinion of the lack of methodological quality of the study and therefore explaining my inability to draw any strong conclusions from this study.
It is important that the literature reviews are presented in a logical, sequential way to tell a story and lead the reader to gain an understanding of the area of research under question. The role of the reviewer is to summarize and evaluate evidence about a topic, pointing out similarities and differences and offering possible explanations for any inconsistencies uncovered.
Conclusion: Though this article, I have attempted to highlight that literature reviews are an essential aspect of design and conduct of any research and help place your research into the wider context. Once you are aware of the landscape of research area in your field, you will be ready to finalise your own research question and research design in preparation for data collection. In the next article of the series, I will explore various research designs and utilize a hypothetical example scenario to present the decision making involved in choosing a research design for your project.