The patient/population of interest may be further clarified by addressing the age, gender, ethnicity, or disorder/procedure/disease in the question. Interventions may include elements such as a therapeutic intervention, a diagnostic test, exposure to disease, or a risk behavior. The comparison is the additional intervention being compared against such as another medication or nursing intervention, another diagnostic test, or quite often routine therapy or standard of care. The outcome of interest is the outcome in which one is interested in seeing a change. Often, one will evaluate multiple outcomes in an EBP project. The most essential component of an outcome is that it is measureable. Outcomes commonly measured in perianesthesia EBP projects include length of stay (LOS) in a particular area, pain scores, incidence of PONV or postdischarge nausea and vomiting (PDNV), and patient satisfaction. Another important outcome measure that should be considered is cost of care.

Once the PICO components have been defined, the next step is to organize them into a question. The most common EBP questions are focused on either an intervention, prognosis/prediction, diagnosis/diagnostic test, or etiology.¹ Templates for developing questions using identified PICO components are provided in Box 9.2. Box 9.3 provides an example of the process.

The process of locating the best evidence is driven by the formulation of a solid question. The PICO components of the question provide the key search terms as well as guides for limiting or narrowing your search. The ideal approach would be to engage the services of a medical librarian familiar with EBP searches. The reality, however, is that such resources may not be readily available to the bedside nurse. The first step to embarking on a successful search strategy is to identify the key search terms unique to your particular question. It is often helpful to begin this process by taking the time to make a list of key search words or terms. The most successful searches are generally driven by at least two or three key search concepts.²² This list should be driven by the PICO components of your question. Using the example question in Box 9.3, key search words would include acupressure and PONV. Other useful search terms may include postoperative, postanesthesia, nausea, vomiting, and complications. Because PONV can occur in both adult and pediatric populations, it may also be helpful to use the adult population as a search limit.

In most databases, these key search terms will automatically map to medical subject headings, also known as MeSH terms.²² For example, if PONV is not a MeSH term in the database you are searching, it may automatically map to nausea, vomiting, or postoperative complications. It is recommended that one use the “explode” option for the primary search terms and then use features such as combining search results using “and” or “or” to further narrow down the results. For instance, should the search term “PONV” map to a MeSH heading of “postoperative complications,” it may be helpful to fully explode this term as well as the terms “nausea” and “vomiting.” One then can combine the three terms (postoperative complications, nausea, vomiting) using “and” to narrow down the search to literature specific to PONV. One would then combine this narrowed search result with the results from the search on the term “acupressure” to capture the literature addressing the use of acupressure for the prevention/treatment of PONV (see Box 9.3). These results may then be narrowed by adult ages using the search limitation options.

If this initial search strategy yields a large number of results that would prohibit a comprehensive review of the references, it may be helpful to further limit the search by levels of evidence. EBP should be guided by the “best” available evidence. What is considered “best” is guided by the level of the evidence and its relationship to the question of interest. The level of evidence is ranked according to the type of evidence or research design. Numerous evidence hierarchies are available in the literature. All commonly rank systematic reviews, meta-analyses, and high-quality evidence-based clinical practice guidelines as the highest level of evidence and expert opinion as the lowest level of evidence. A sample evidence hierarchy is provided in Table 9.3.