In this step you critically appraise the evidence for its validity, impact, and usefulness in clinical practice. If you do not use a systematic reviews database, you will have to appraise whether or not the research is clinically sound. There are many different types of research studies, including prognosis, diagnosis, harm/etiology, and therapy. You will need to examine the articles on the basis of validity by examining potential sources of bias, as well as looking at the type of methodology used in the study. For more information about these, see UNC's HSL Evidence Based Medicine tutorial and the readings below. Regardless of whether you use CINAHL or Cochrane, you will need to assess whether the articles you are reading are appropriate to your clinical situation and patient's needs.
Read the following articles, looking for potential sources of bias, examining the methodology, and validity. Compare the usefulness of the articles to your question: Will decreasing the number of RNs (versus UAPs) have a low impact on patient care?