Library Services for International Learners at UHN: Search Results

Ask yourself if the number of citations are managable for you considering what you need them for and how soon the project in question needs to be completed.

Consider:

Bringing your search to your Information Specialist - we can help you get better results!

Eliminating some of your keywords or subject headings. 

Figure out which ones are bringing in the most useful and the least useful results, and get rid of the least useful ones.

Adding another concept. 

Maybe [depressive disorder] AND [drug therapy] is too big (over 5,000 in MEDLINE). 

If you go back to your PICO and look at the columns you didn't search from - outcomes, for instance - you may get better/smaller results. 

In this case searching [depressive disorder] AND [drug therapy] AND [Sleep Initiation and Maintenance Disorders] could narrow your search to some side effects you are interested in, and brings the results down to under 100 articles.

Adding or tightening filters. 

Think about the question you are trying to answer. 

• Does it make sense to only look at things published in the last 5 years?  10? 
• Age groups can be a great way to cut out irrelevant material. 
• You can add a Clinical Queries or EBM filter - in this case one of the Therapy ones. If you already had one on, you can tighten it up, for instance by switching to the "maximizes specificity" version instead of the "best balance" version. 
• If you are especially interested in one or two types of papers, you can add a publication type filter. 

In the example we're using, adding a "therapy, best balance" filter and limiting to 5 years gets us down to under 20 results.

Once your search results are managable, your first step is to scan the citations for relevance 

Read the titles - do they address your PICO or research question? If no, discard.  If you're not sure, set it aside. 

Once you've gone through the titles, read the introductions and conclusions of the abstracts for the citations you haven't discarded.  Do they address your PICO or research question?  If no, discard.

Be careful not to get caught in "scope creep" - try not to put a citation in the yes or maybe category just because it is about something similar, related or interesting.  It needs to address the question you are asking.

If, at the end of this, every citation is marked NO, consider:

• expanding your search - maybe some of the limits cut out articles you could have used.  This may mean going through more citations.
• changing the terms you used - maybe there is a better, more specific subject heading and/or synonym for you to use
• switching databases.  Maybe MEDLINE doesn't have anything on your topic but PsycINFO or EMBASE does
• finding a key (a useful paper, guideline, review, researcher name, textbook, etc. that you already know exists) and using the citations, authors, terms, etc. that you find through them to search

Once you have decided that an article is relevent, you need to decide whether it is valid.  The books and links below will help you evaluate the methods the researchers used, and the conclusions that they came to: