Who is this for?
If you are looking to do a literature search for a particular essay/dissertation, or you are wanting to research a particular area, this guide is for you. If in doubt, it may help to work through the basic guide to finding resources first.
Before you jump onto a computer
Planning your search thoroughly is vital to conducting a good search that pulls up relevant resources. In the basic guide we talked about identifying key concepts from an essay question. Now we want to take that a step further and identify our search terms in a systematic way.
There are different methods for approaching this but we are going to focus on just two here: one specific to health (PICOT) and another which could be used to research any topic (Concept Boxes). Details of each are below but do be aware that different methods are out there…
The first method for breaking your research topic down into concepts is PICOT (for the purposes of searching we will not use the T), which is used for searches that are focused on clinical research. Let’s break down what each element of the PICO acronym refers to:
P – Patient / Population / Problem
How would you describe the group of patients you are interested in? What are the most important characteristics of the patients, which define them as a group?
I – Issue / Intervention
Which is the main prognostic factor, intervention, treatment, or exposure you are interested in?
C – Comparison
What is the main alternative to compare with the intervention? You may not need to include this in a search strategy as it is more relevant to those developing research questions – always consider it though.
O – Outcome
What outcome are you interested in? Are you looking for a specific outcome, or are you looking at a number of different outcomes?
Now that we have defined the PICO method of breaking down a research topic, let’s look at an example of it in action.
Let’s imagine we are researching the following question:
Does the use of SSRIs in the treatment of adolescents with depression reduce suicide rates?
Using the PICO method, we’d organise the key concepts as follows:
|Adolescents with Depression||SSRIs||n/a||Suicide rates|
Once you have done this, remember that you need to come up with as many different ways of describing each concept as possible. For example, for adolescents we might also search for teenagers or young people or young adults, etc. For SSRIs, you may want to search for specific drug names too and so on.
For a recap on the PICO method, watch this video from the Bodleian Library:
Concept boxes or concept mapping is a more generic way of breaking down your topic. Here, rather than using prescribed groupings like with PICO, you will simply be pulling out key concepts from your topic and organising them using your own principle (often by importance / centrality to your topic). Again, let’s look at an example to illustrate this.
Imagine we were investigating the following question:
Discuss the treatment options to alleviate severe angina in older patients.
Using the concept boxes, we’d break this question down as follows:
Notice here that the headings for each group are generic and that those concepts I have deemed most important have been listed first. Remember, once you have determined the key concepts you are interested in, you need to come up with as many ways of describing each group as possible.
Give it a go
Now that we’ve looked at two ways of organising the key concepts you would like to look at, try analysing the two examples below using both methods – in each case, think about which method was more useful and compare your groupings with mine by clicking ‘Check the answer’. Do you agree with what I’ve done or could I have done anything differently?
Discuss reasons for the under-representation of older patients from south-east Asian backgrounds presenting eating disorders.
Discuss the links between severe alcoholism and respiratory problems.
Hopefully these two methods have given you a start when it comes to thinking about how to break down your topic. There are other methods too, and you may like to read about methodologies in more detail. The Library has some good books about this and I can recommend Doing A Systematic Review or Doing Your Literature Review as good places to start.
Once you’ve broken down your topic into concepts, you need to think about different ways each concept could be described and make a list of terms for searching. If you don’t feel confident in doing this, go back to the basic guide to searching and read the section on finding journal articles, which will walk you through how you ought to generate lists of words to search for.
Going beyond LibSearch
When it comes to conducting a good literature review, you will need to go beyond LibSearch and use subject-specific databases where you can access better information, which has been categorised for your discipline specifically.
You can view the full list of databases we subscribe to at UEA by visiting the Find Databases page. However, academics in your departments – notably your supervisor – may suggest databases they consider most relevant for your particular topic. To get you started, here is a list of good databases that may be useful (note, only pick the databases most relevant to your topic – you don’t need to search them all):
An index database. Contains journal citations and abstracts for biomedical literature from around the world from journals covering medicine, nursing, dentistry, veterinary medicine, health care systems and the biomedical sciences. Available via 3 alternative routes:
There is no difference in the Medline content available from any of these sources – the only thing that differs is the interface used to search the content. To get you started with one of the options, here is a video guide to using PubMed:
An index database of English-language and selected other-language journal article citations about nursing, allied health, biomedicine and healthcare.
Here’s a video from the University of Sydney to get you started:
An index database. Contains journal citations and abstracts for biomedical literature with over 28 million indexed records from thousands of peer-reviewed journals and conference proceedings — includes over 6 million records that can’t be found in MEDLINE.
Here’s a video to help get you started:
The Cochrane Library is a collection of six databases that contain different types of high-quality, independent evidence to inform healthcare decision-making, and a seventh database that provides information about groups in The Cochrane Collaboration.
Here’s a video to get you started:
PsycINFO® is an expansive abstracting and indexing database with more than 3 million records devoted to peer-reviewed literature in the behavioural sciences and mental health. Full text of all APA journals is available through this database.
Here’s a video to get you started:
Social Care Online
Social Care Online is produced by the Social Care Institute for Excellence (SCIE) and is the UK’s largest database of information on all aspects of social care and social work.
If you’ve used any advanced databases before, you’ll know that the majority work in a very similar way so once you understand the concepts of building a good search, you’ll be able to query any database like a pro.
Basic Search vs. Advanced Search
Pretty much all the databases you access will have both a basic search and an advanced search option. The basic search will look a lot like a Google search bar where you can pop in keywords and hit search nice and quickly. They will normally look something like this:
We are not going to use the basic search function, however. For us, the advanced search will be much more useful and allow us to build a strong search string more easily. Often you will see an option for the advanced search near the basic search so all you’ll need to do is click on this to access the more powerful search tools. An advanced search will look something like this:
Different databases will all look slightly different but most will have the same core functions as shown above. The key thing to look out for is multiple search boxes rather than just one (or the option to add more search boxes by clicking a + button to the right of existing boxes). Where this is the case, you can be confident that you are in the advanced search area.
Using Advanced Search
As advanced search tools give you an unlimited number of search boxes, we will be able to organise our searches really neatly. What we will do is use a separate search box for each of the core concepts we have identified in our PICO (or alternative) analysis of the topic area. Let’s use our example from above to demonstrate:
If we were going to run an advanced search for these concepts we’d map them to the search boxes as follows:
Now, this is a good start but you will have developed more than one search term for each concept. For example, to cover adolescents, you might also need to search for young adults, teenagers, young people, etc. That is easily done using an advanced search. All you need to do, is add each of these words or phrases into the search box which relates to that particular concept, and separate them with an OR. For example:
This means you will then be search for papers that contain any of the words or phrases you’ve selected for this concept.
There is one final thing to do before we are finished with this concept. You will see that the drop down next to our search bar currently says ‘All fields’, this means that when the search is run, articles that feature one of the selected words anywhere in their record of full text will be returned as a result. That may be ok, but when you are looking for a key concept you may want to reduce the likelihood of bringing back results that aren’t relevant. To do this, click the dropdown menu attached to the search box to see more options.
You will get slightly different options on different databases but the two most useful/common options normally available to you are: Title and Abstract. If you select either of these options, the search will only include articles that feature one of your search terms in its title or abstract (dependent on which you select). This should indicate that the concept is very important to the article and therefore relevant for you.
It is worth being cautious about running Title-level searches in particular as articles are often titled provocatively to catch interest and may not feature important terms as a consequence.
A useful way to navigate this potential problem, is to make use of MeSH terms (not confident using MeSH? View the MeSH guide). MeSH is a hierarchically-organised list of terms that relate to the area of medicine and health. In short, it is an agreed set of words that can be used to describe different aspects of the subject area. If you’re interested, have a look at the full MeSH index (useful for finding particular terms).
Articles that appear in some of the biggest databases will have been checked by experts and correct MeSH terms attached to them to indicate what the article relates to. For example, the MeSH term for suicide is (unsurprisingly) suicide. So we could use that instead of our original phrase suicide rates to pick up any articles relating to suicide.
If we continue with our example, the search will now look more like this:
There is no one right way to set up a search – here I’ve searched two concepts at Abstract-level and the other using its MeSH term. Do you agree, would you have done something differently?
The final thing to note is that between each new box, we have a drop-down menu which defaults to ‘AND’.
By leaving this as AND, we are telling the database that we want to view articles which contain any of our words/phrases for concept one (adolescents) AND concept two (ssris) AND concept three (suicide). So we should only get back results that cover all three of these concepts – that’s searching done well!
A final note on this, you can also switch AND to NOT, which may be useful if there is a particular type of article you don’t want to look at. For example, in our example we are interested in suicide rates so we may want to build an additional concept box that covers suicidal or self-harming behaviours which don’t result in suicide itself. If we did this, our search would finish up looking like this:
When you get to this point, the only thing left to do is run the search and check over the results. You may find that you have to play around with search terms and the way you structure your search so don’t be afraid to adapt and definitely don’t try to develop a perfect strategy that will work first time. Each new search is a new learning process so think about what you’re doing and why you are doing it. If the results of a search aren’t what you were hoping, have a look at your terms and make sure they are relevant and not too general, that you have thought through all the different elements we’ve talked about above, and if all else fails try breaking your search down into its different concepts and running a search for each individually to try to understand which elements of your search are working well and which aren’t.
You should have all the information you need to run some supercharged searches now but if you run into any problems, remember, you can always come to the Library and ask for extra help – that’s what we’re here for!