1. Module Five Search Demonstration
In this module we are going to walk you through a hands-on demonstration of a search and demonstrate how to use some of the tools that we have discussed so far.

2. Electronic Demonstration of Search
Search Strategy
Start with meta-analyses and more exclusive search engines (e.g. Campbell Collaboration and EBM Reviews)
Consider articles or reviews that are “close” to the topic when there are no exact hits
Be sure that you are sitting at a computer with an internet connection and that you have your search terms worksheet filled in with key search terms, your MOLES worksheet, your EBP tools handout, and the example search tracking form. What we are going to do is walk through an example using a COPES question that was chosen by a group of practitioners who participated in this training. The question that they were trying to address was: “If young immigrant couples who have a history of difficult ‘attachments’ (experienced a history of disrupted relationships in their family of origin) are treated together as a couple (as in marital or couple’s counseling), or are given intensive individual counseling, which will result in: 1) fewer instances of family violence 2) increased parenting skills 3) fewer child behavioral problems.” For simplification we will only focus on one effectiveness outcome: parenting skills.
If you remember, we suggested that you start with the top of the pyramid: systematic reviews and meta-analyses. For this reason, we will begin our search using the Campbell Collaboration website. First we will scan the library for systematic reviews that might address our question under the “social welfare” heading:
We don’t find any that specifically address our question, but there are several treatments related to parenting skills generally. If we are unsuccessful at answering our particular question, we may want to visit these related systematic reviews and consider reviewing those treatments. For now, let us continue with our current search.

3. Search Tips
Move on to less restrictive or more general databases if desired studies are not found
Boolean Operators
Truncation
Keep track of terms
We will move to another engine on the Campbell site, the C2-SPECTR advanced search:
Each search engine is unique. In this case, all of the studies included are randomized trials, thereferor the MOLES search terms which are meant to capture randomized trials are not necessary to use here. Instead we will use some of our clinical terms to search.
It’s always a good idea to read over the “help” or “faq” sections of a website to learn about how to maximize your search. Many search engines use “Boolean operators” such as “and” “or” and “not”. If you are unfamiliar with the use of these terms, you can visit the Campbell’s help site under “advanced search”:
Now, take a look at the search tracking example. You will want to keep track of your future searches so that you don’t search the same terms twice and you can retrace your search if necessary. You’ll see we’ve recorded the Campbell Collaboration website in the first column, then listed the clinical search terms that we’re going to use in the second column. Go ahead and type these terms in the C2-SPECTR advanced search website on your computer. Pay careful attention to when we use “and” versus “or”. We used “and” between “couple” and “individual” because we want a study that compares both.
We see that only two studies appear to meet our search term criteria. You may find more studies when you try this same search as studies are constantly being added to the body of knowledge available. The two studies that we identified with this search don’t appear to fit. One is focused on alcohol problems, and the other is focused on chronically ill children. The search we used might have been too narrow.

4. Try several different terms Art of Trial and Error
More Search Tips
Try several different terms
Art of Trial and Error
Record the full reference of potentially helpful articles
Get to know a few databases really well
Now we will adjust our search in order to be more general. Perhaps we can use other, more general terms such as “therapy” and “treatment” to address our outcome of interest: parenting skills. Go ahead and type in these terms on your computer.
We find 126 “hits” this way. So, being more general with our terms gives us considerably more potential studies to inform our question. These results can be scanned by clicking on the “view” button to the right of each title. You can move through the results by clicking on the “page” button in the upper right hand corner.
Although no study directly answers our question of interest, some titles do suggest promising treatment options. For example, titles of two studies include “Problem-solving skills in happy and distressed couples: effects of videotape and verbal feedback” and “Effects of behavioral marital therapy on couples' communication and problem-solving skills”. The outcome seems to be “problem solving” rather than “parenting” skills, but these studies could be useful to consider. Be sure to record the full reference for studies that answer your question, or may be helpful. The Campbell registry provides abstracts, but not the full article text, of studies so you’ll want to keep a running list for yourself.
Practice navigating the Campbell website on your own. If you were to continue searching on this example question you would try several different search terms in order to identify the best studies. You may want to try some search terms related to your COPES question at this time. The idea is that you will become increasingly familiar with and skilled in using a few databases. Let’s try searching on a different database.

5. Using MOLES Truncation List all terms in the MOLES column Keep trying
We will now move our search to the PubMed database. This time we will use our MOLES terms for effectiveness type questions. You’ll see on the example sheet that we entered in all of the terms from the MOLES sheet under effectiveness as well as the terms “parent*” and “skills”.
As we said earlier, it’s always a good idea to read over the “help” or “faq” sections of a website to learn about how to maximize your search. Aside from Boolean operators, you’ll also want to understand the truncation tool if it is available. This corresponds to the “*” on your moles sheet. Truncation could be another symbol, such as the “$” used by the EBM reviews website. If you are unfamiliar with truncation tools, you can view a description on PubMed’s website:
Go ahead and enter the terms listed on the example worksheet on your computer including the “*” symbol. When we performed this search we came up with 2179 hits- you may have an even higher number. This is far too many to scan. We need to get more specific with our search terms. Right now the only clinical terms we are using are therapy, treatment, parenting, and skills. As you see, using the same terms in one database and then another can result in drastically different results. Part of the reason is that some databases, such as PubMed are much larger and contain many different types of articles and studies. Let’s try to make our search more specific by adding the term “immigrant*” to our search. Go ahead and enter this search on your computer.
This time we get only 3 hits, and none of them quite fit our question. As you can see, many different search terms and attempts may be necessary to identify individual studies. This is part of the reason why it is much more preferable to find systematic reviews where someone else has already searched out and synthesized individual studies. At this time you may want to try searching some of the terms related to your COPES question using the PubMed database. Remember to track what you find and record the full titles of references that you find that might be helpful in answering your question.

6. Other Potentially Helpful Sources of Research Evidence
Experts in the field
Google Scholar
References in other articles
Although many databases are available to you on the internet, you will want to consider other sources of information or evidence about your question. If you can identify an expert researcher in your field, perhaps at a conference or from a local university, who would be willing to discuss your question with you, they may be able to save you some time in searching. This can be an excellent first step in exploring your COPES question.
If you are coming up with very few studies that address your question you may want to try Google Scholar or other internet sources. Keep in mind that the quality of information on the web varies greatly. This is a good way to search out new terms in an unfamiliar area, but should not be used as your primary source of evidence for decision making.
Another helpful hint is to look at the references that are cited in one article for other studies or articles that might be helpful to you. Researchers will often summarize the state of the art in a particular area of research in the literature review section of their manuscripts, which can lead you to other research in the area.

7. Team Search Plan
Who will be “searchers” (the whole team, one or two people)?
How to divide the search (by outcomes, search terms, databases etc.)?
How many articles would we like to bring in?
If you are working through this training as a team, now is the time to discuss your search strategy. If you are working alone, you can move on to the last slide.
Some people may feel more comfortable using search tools, databases, and engines than others. If you are going to divide the search responsibilities among team members plan ahead in terms of who will search for which terms in which databases so that work will not be duplicated. Also, make a target goal for yourselves in terms of the number of citations you would like to identify. For the purposes of this training, we would like to keep these goals manageable. However, in practice, there is no set limit to the number of pieces of “evidence” that should be identified. Some questions may produce great amounts of research evidence, while others do not. Keep in mind that you may look to evidence that is related to, but does not exactly address, your question of interest when perfectly matched studies do not exist.

8. Wrap-Up For next time: practice searching
Make a list of any interesting articles, studies, meta-analyses, etc.
Next time: judging quality
At this point in the training you should take time to practice your searching skills. Be sure to read the help or faq sections of the websites you choose, try different databases, and use many different search terms. Keep track of your search and make a running list of references. You may want to include abstracts as well, particularly if you are working as a group, so that you can share the evidence you find with your team members.
In the next module we will discuss judging the quality of evidence.