Informatics in Health Care

PICO: A Model for Evidence-Based Research Video Transcript

This tutorial will teach you how to use PICO, a model that outlines your clinical research question. It acts as a framework, asking you to think specifically about different aspects of what you want to investigate, and is a fundamentally part of evidence-based research. The first four letters are generally always needed, while the two Ts on the end may be optional. Let's start with the first four.

P is for patient or population. Who is receiving the treatment, and what are their characteristics? I is for intervention or interest. What do you plan to do, such as a specific treatment or therapy? C is for comparison. What are alternative methods or treatments to your intervention. And O is for outcome. What do you expect to see or want to happen? We'll now go into a little more detail on what each letter means.

Patient or population. How would you describe your patient? What are the most important characteristics of the patient? This may include the primary problem, disease, or coexisting conditions. Gender, age, ethnicity, or location are additional descriptors that may be relevant to your research question. Intervention or interest. What main intervention, prognostic factor, diagnostic test, or exposure are you considering? What do you want to do for the patient? Prescribe a drug, perform a test, order surgery? Comparison. A good analogy for comparison are the variables and controls in a scientific experiment. The variable is the group you are experimenting on, and the control is the group you are leaving alone to compare your variable with. The comparison in PICO can be thought of as the control group. You're comparing your intervention with another group or alternative. Are you trying to decide between two drugs, a drug and no medication, or placebo, or two diagnostic tests? Outcome. What are you expecting to see? What are you hypothesizing that your intervention will cause?

Time. Is there a timeframe? How long will it take for the intervention to achieve an outcome, or how long will participants be observed? Are you following women during pregnancy, observing sport injuries over the season, following study participants for a set recovery time? Type of question or study. There are several different designs or methodologies of studies. You may need to think about which type will best fit your question and the quality of evidence you are most interested in or likely to find for your specific question.

Now that we know what PICO stands for, we'll be going through an example. Our initial research question is, "What effect will the campus smoking ban have on students who smoke?" Think about the question with the PICO framework in mind. What are some ways to describe the patient, intervention, comparison, and outcome? Our patients are students who smoke. We can also describe them by location, which is college campus. The intervention would be the smoking ban, and a comparison could be not having a smoking ban. One outcome we could hypothesize is that students will smoke less. You may be thinking of a different outcome. We'll go with the smoking less outcome for this particular example. You may want to include a timeframe, such as a semester, or even a type of study. We'll skip those for now.

Using PICO, our question is now, "Do college campus smoking bans decrease the amount that students smoke?" Now that we have our basic PICO, we want to identify the keywords we can use to search for resources. To start, we'll identify the main concepts of our question. Students will definitely be a main concept of our search, as well as smoke and college. We probably don't need to consider the word campus when it comes to actually searching for literature. Smoking is obviously a main concept, similar to smoke. The word ban is also going to play an important role. Finally, smoking less is a concept we want to be sure to include in our search. Take a moment now to think about possible additional keywords in each PICO category. These can be synonyms or just different ways to describe the words we already have. This will help us think more about our topic, and help our searching in order to find the best, most relevant results.

Starting with students, the first thing we may realize is that we are interested in both the plural and singular form of the word, as is the case with some of our other words. When thinking about colleges, we can also use the word university to describe the same concept. When it comes to smoke or smoking, we may also be interested in tobacco. There are a lot of ways to talk about a ban. Regulations, laws, and the phrase "smoke-free" are a few examples. Lastly, we have the idea of smoking less. The key part of this is less, since we already have words for smoking. Other ways to express this concept are quitting, reducing, or decreasing. Now we have a good set of words we can begin searching with. We can group these words together into four main concepts: our population, students, the intervention, banning or regulation of smoking, smoking itself, and the outcome of decreased smoking. It's helpful to keep a document of your concepts and words or phrases for those concepts. This will help you as you search, and you will likely find yourself adding to the list as you explore your topic further.

Our last step is to turn this into database language. Databases use what is known as bullion operators to further define a search. These are and, or, and not. And “and, or” are the key operators to use. And is going to combine all of our concepts together. This is our way of telling the databases that all of these concepts need to be included in the results. We don't just want results talking about smoking in college. It also has to talk about a decrease, or lack thereof, in smoking, and some sort of regulation in regards to smoking. Or is the way to tell the databases that we are flexible. We would use or to combine all or some of our words within each concept. We don't really care if the database finds items using the words college, or students, or university. It just has to use at least one of them.

You may have noticed that some of our words are very similar to each other, like smoke and smoking. In this example, we are or-ing them together, but it can be tedious to have to type both words out in our search. Don't worry, though. There are tricks. Instead of searching both words, we can search using truncation, by putting an asterisk at the end or even the middle of a set of letters. This represents a wildcard, so any letters appearing after the asterisk are included in the search. If we remove the different endings of the word and add an asterisk, we are telling the database to look for the core word no matter the ending. Our results will now bring back smoke, smokes, smoking, smoker, and even just smok by itself, even if it's probably a typo. This is also useful with plurals. Finally, you may have noticed that we put the phrase "smoke-free" in quotes. This is another database trick which tells the database that these words have to be found together. Quotes are extremely useful for phrases or specific titles.

Now we've just covered the basics of PICO, and you're set to use it on your own. If you need further assistance, contact a librarian using the Ask a Librarian link at the top of any library's page.