Once a clinical question or problem has been identified, whether by a nurse at the bedside or by a team within the institution, the identification and revelation of the evidence related to the problem is the next step in finding a solution to the problem. There has been much discussion about what type of information can be considered as evidence. Evidence can range from very scientific research publications such as RCTs, survey research, and cohort studies to case studies of interesting or unusual cases, expert opinions, or consensus conference recommendations. 12.^{26.29. and 36.} RCTs are generally large experimental research studies designed to randomly assign patients to experimental and nonexperimental groups and compare the results of one form of treatment against a control group, which generally receives the current standard of care. The researchers then determine if the new experimental treatment is more effective than the current treatment. The “gold standard” for evidence used to guide practice is clearly RCTs, systematic reviews, and meta-analyses. ^{29.32.33. and 38.} Systematic reviews and meta-analyses are types of research whereby the authors review multiple studies on a particular topic, separate the best research, critically analyze the research, and then make practice recommendations based on the findings. ^{17.33. and 38.} These very carefully and systematically designed research trials have the greatest ability to inform clinicians about the best practices based on research. It is unrealistic to think that individual nurses or physicians have the skill, knowledge, or time to conduct systematic reviews of every question they have regarding their practice³⁸; therefore systematic reviews or meta-analyses can be very useful to the health care provider. The Cochrane Collaboration is an international organization that was initiated in an effort to promote the accessibility of systematic reviews on the effects of health care interventions. ¹⁴ When a topic is chosen by the Cochrane Collaboration to undergo a systematic analysis, a rigorous standardized review process is undertaken by a group of experts on the topic. Once the review is completed, recommendations for practice are published. Cochrane reviews have been found to be more rigorous than either systematic reviews or meta-analyses published in peer-reviewed journals. ¹³ For more information on interpretation of systematic reviews, the journal Archives of Pediatric & Adolescent Medicine has published an excellent three-part guide for interpreting systematic reviews. ^{15.17. and 20.}

Although RCTs, systematic reviews, and meta-analyses are the gold standard for providing evidence, many topics do not have an RCT, systematic review, or meta-analysis of the data available to help guide the clinician. Therefore it is important for the nurse to examine the available data on cohort studies, smaller experimental studies, qualitative studies, or case studies. Although it is clear that this type of research represents a less rigorous research design and is rated lower on the hierarchy of evidence, it is still very important information that will alert the nurse to trends or problems in current practice that need to be identified and examined. Most large trials result from a compilation of smaller studies that alert the scientists to the need for larger studies. Case studies are excellent means of highlighting a problem or issue that needs to be addressed by the health care community. A case study generally highlights an unusual or interesting case a practitioner has encountered. Although this is not considered scientific evidence, it may represent an interesting finding or a future trend in health care.

Conducting a literature review takes time, a great deal of research knowledge, and mastery of the art of reviews. Many health care organizations employ research librarians who can be an invaluable resource to nurses conducting a search for evidence. The review must be conducted in an organized and systematic manner. The first step of the literature review is to formulate a question that provides clear direction for the literature search. For example, What is the best treatment for croup? can clearly be researched, as opposed to a question that is nebulous and difficult to search such as, How can patients live with croup?

Once the question has been formulated, a search can begin. The first step is to conduct a computer-generated search on the keywords involved in the topic. Common keywords can often be found within databases. Common databases used by nurse researchers include CINAHL (Cumulative Index to Nursing and Allied Health Literature), MEDLINE (Medical Literature Analysis and Retrieval System Online), CANCERLIT (cancer literature), and PsycINFO (psychology information). ^{26. and 33.}

When beginning to conduct a literature search, a nurse must first understand the difference between primary research and secondary sources. A simplified explanation of these two concepts is that primary research studies are studies published by the researchers who actually conducted the study. Secondary sources are authors who publish papers about research conducted by others. Because secondary sources generally report, critique, and evaluate the research of primary studies, they are providing their interpretation of the research and do not provide a detailed account of the original research. ²⁶

Finding the best research or secondary sources that evaluate the research takes skill and practice. It is not good enough to simply find all the research and literature published on a specific topic; it is necessary to find “evidence that matters.” Evidence that matters refers to finding evidence that has implications to change a patient’s “overall prognosis and outcome and quality of life.”³⁸ An acronym exemplifying this concept was coined in 1994 by Slawson and Shaughnessey—POEM, patient-oriented evidence that matters. ³⁸ This concept was developed by a group of clinicians in England that publishes monthly findings of POEM research. They also have an active web site (http://www.infopoems.com) with monthly updates on current reviews. Before the evidence on a particular topic can be considered for a possible change in practice, the evidence has to meet the following three criteria (http://www.infopoems.com/index.cfm):

Searching for systematic reviews can also be conducted using a search of systematic review sources. Some of the sources that publish systematic reviews are listed in Table 5-1 and are available online, either free or with a yearly subscription. Journals can also be a rich source of systematic reviews, and a list of journals dedicated to publishing systematic reviews is included in Table 5-1. ^{22.26.33. and 38.}