Evidence-Based Medicine and Epidemiology
DEFINITION: EVIDENCE-BASED MEDICINE
Evidence-based medicine (EBM) is “the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients” (www.cochrane.org). It is also known as evidence-based practice. There will be several questions on the American Nurses Credentialing Center (ANCC) exam that will test your ability to sort and rate articles by the level of evidence.
HIERARCHY OF RESEARCH EVIDENCE (FIGURE 29.1)
Figure 29.1 Ranking of research evidence. This image can be found in color in the app.
This is a statistical method that combines data from multiple studies (systematic review), resulting in higher statistical power and a single conclusion. This method is considered the gold standard for gathering research evidence for EBM.
A type of literature review that identifies, selects, and analyzes multiple research articles concerning a health condition, disease, or other health-related practice. Follows specific methodology to identify all the relevant studies on a specific topic. Studies to be included must meet explicit criteria. Studies are ranked from grade A (best evidence) to grade D (poor evidence). After a systematic review is done, the acceptable studies are pooled together and statistical testing of the data (meta-analysis) is performed.
Randomized Controlled Trial (RCT)
Subjects are randomly assigned to either the control group or the treatment group(s). The intervention may be a drug, procedure, or device. Some RCTs use a double-blind design (the intervention is hidden from the patient, clinician, and/or researchers). RCTs are experimental studies.
In a nutshell, an experiment involves random subject selection, one placebo or control group, and one or more intervention group(s). An RCT is a type of experimental study.
Cohort studies are a type of research that is used to investigate risk factors for diseases, risk factors for death, and other conditions. The research subjects are observed for a long period. There is no intervention done (not an experiment). The goal is to identify risk factors and associations (not causation) of a disease(s). For example, the Nurses’ Health Study is a large cohort study of female RNs aged 30 to 63 years who reside in the state of Massachusetts. A cohort study can be a type of prospective study (present to future).
A detailed report of one patient with a disease or an unusual condition that includes demographics, signs and symptoms, diagnosis, response to treatment, and so forth.
A series of case reports that involves a series of individuals who are given similar treatment.
Opinions and Editorials
Opinions and editorials can be biased and may not be based on solid evidence. They are the weakest form of evidence.
The gold standard database and resource for EBM. These are systematic reviews (Cochrane Database of Systematic Reviews). The organization does not accept commercial or conflicted funding. Also known as the Cochrane Collaboration (www.cochrane.org).
The U.S. National Library of Medicine (NLM) premier database contains more than 23 million journal articles in the life sciences with a concentration in biomedicine. These articles are from 5,600 current biomedical journals from the United States and more than 80 foreign countries.
This component of MEDLINE contains over 27 million citations of biomedical, medical, and other life science literature and abstracts.
Cumulative Index to Nursing and Allied Health Literature (CINAHL)
The world’s largest source of full-text nursing and allied health journals (>1,300 journals). CINAHL Complete provides indexing of more than 4,000 journals.
GRADES OF RESEARCH EVIDENCE
Research evidence receives a letter grade: A (best evidence), B, C, and D (poor evidence). Well-designed controlled experimental trials (double-blind RCTs) are considered to be grade A (or level 1) evidence.
HOW TO SOLVE EBM QUESTIONS (DRAG-AND-DROP FORMAT)
1. What is the “drag-and-drop” format in the American Nurses Credentialing Center (ANCC) exam?
This is one of the new-format questions. On the left side are three research articles/studies (white boxes are shown here, but they will be yellow on the exam). On the right side are three boxes numbered “1, 2, 3” (black here, but blue on the exam). You must rate the three articles in terms of evidence from best (1), moderate (2), and worst (3).
2. What is the best way to answer this type of question?
The easiest way to answer this type of question is to memorize and understand the highest level of evidence (meta-analysis, systematic review, RCT) and the type of studies that has the lowest level of evidence (opinions, editorials). The leftover study belongs in the middle black box (#2). The first sentence of each article usually gives a clue about the study design. Notice that in option A it is a “meta-analysis,” in option B it is a “specialty society opinion,” and option C it is an “experimental study.”
I teach the following method to my review-course students. These are the steps to follow:
1. Identify the article with the strongest level of evidence (#1 ranking). Look for key words such as “meta-analysis,” “systematic review,” “RCTs,” and the Cochrane, MEDLINE, and/or CINAHL databases.
2. Next, look for the research study that has the weakest evidence (#3 ranking). It has key words such as “expert opinion,” “opinion,” or “editorial.”
3. You are left with one article, which you drag to the middle (#2 ranking).