The first step to finding the information you need is to ask a good question. This can be harder than it sounds! There are, however, three key reasons to do this:

1. Focuses the question by identifying the essential components
2. Defines the concepts to be used when searching a database
3. Helps to identify what kind of information (research articles, book chapters, policy examples, etc.) will answer your question

There are lots of strategies you can use to develop a good question. One popular one in nursing is called the PICO method. The letters in PICO stand for specific elements of a well developed question:

• P - Patient/Population/Problem

• I - Intervention

• C - Comparison/Control

• O - Outcome

Here's an example of how you could apply the PICO format to a specific research topic:

Chart based on example from http://www.cebm.net/asking-focused-questions/

Creating a PICO question can also help you figure out what keywords you should use when searching a database like CINAHL or Google Scholar. Simply look at the concepts in each component of your PICO statement and pull out a word or two for each one that clearly and concisely represents that concept.

Here are the concepts from our example about phototherapy for newborns with jaundice:

Things to keep in mind:

• Computers are pretty dumb and they won't look for synonyms for a term that you search. That's why you have to think of them. In our example, we would search for both "light therapy" OR "phototherapy because they both mean pretty much the same thing.
• When it comes to the "P" concept, you would usually search for problem element - in this example it's "jaundice" - and save the population part. The reason why is because most databases have limits that make it easy to zero in on groups of people like newborns.
• You will probably not include all the concepts you've identified in your first search. Normally, you'd use the "P" and the "I" concepts and see if you get any good results. If you don't get anything, you can try mixing in the other elements, usually the "O" to see if that pulls anything up.

After you have written your question, it can be helpful to stop and think about what kind of question it is. Our example about light therapy for newborns with jaundice would be considered a therapy question, because it is trying to determine if light therapy is better or worse than no light therapy.

Determining the type of question you are looking for can help you figure out what kind of study you should target. As the chart below demonstrates, randomized control trials, case controlled studies, or cohort studies are usually the best type of evidence to use. This can help you figure out where to search for information on your topic.

Chart based on example from Duke Introduction to Evidence Based Practice tutorial - http://guides.mclibrary.duke.edu/c.php?g=158201&p=1036067