Quickly and Easily Identify Low-Quality Articles Using CRAAP
This is an excerpt from Evidence-Based Practice in Athletic Training by Scot Raab & Deborah Craig.
Chapter 1 outlined the following five steps of evidence-based practice:
- Create aclinically relevant and searchable question concerning your athlete.
- Conduct a database search to find the best evidence.
- Critically appraise the articles or evidence for quality.
- Critically synthesize the evidence to choose and implement a treatment.
- Assess the outcomes by monitoring the athlete.
Chapter 2 demonstrated how to perform a database search, and chapter 3 introduced ways to understand and evaluate the evidence. This chapter continues the discussion of evaluation methods by addressing how to appraise the diagnostic articles you have found. Following are four steps to appraising diagnostic research:1
- Determine the applicability to your athlete.
- Assess the quality of the article.
- Review the results.
- Summarize the clinical bottom line.
As you gather your articles using a variety of search methods, you can apply a simple assessment to quickly and easily eliminate low-quality articles. Review any article based on the following five criteria (CRAAP):
- Currency. Is the information in need of being revised? The 2004 National Athletic Trainers' Association (NATA) position statement on the management of sport-related concussion allowed for same-day return to participation if symptoms resolved and the athlete passed a return-to-play test.3 The 2014 position statement update noted that our knowledge about how to treat concussions had significantly changed and ruled guidelines as recent as 10 years old obsolete.
- Relevance. Is the article written at a level appropriate for your needs, and do you understand it? If not, you may fail to reach a valid conclusion about the use of assessments in your situation. Has the information been peer reviewed, or is it an editorial or opinion piece? Editorials often have a bias and lack the scientific rigor required to make the information useful for sound clinical decisions.
- Authority. Is a sponsor involved that has a product to market? If so, read the information with great care. If an article is promoting an aquatics rehab program to aid in the recovery of knee surgery, but it was sponsored by a company that sells therapy pools, you need to suspect some potential bias. These types of conflicts of interest can be questionable.
- Accuracy. After completing your primary review, it is time to dig a little deeper and appraise the article. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS) is a 14-question instrument to assist you in the appraisal (see the sidebar).1,4 The answer choices are yes, no, or uncertain; a greater number of yes answers is preferred.1 There is no set number of acceptable yes answers, but if any uncertain or no answers make you feel uncomfortable using the data to make a clinical decision, you may want to further research the topic and expand the number of articles you review.
- Purpose. Why is the information presented? Is it an advertisement? Is it funded by someone with a financial stake in selling something? If the answer to the preceding questions is yes, be sure to thoroughly review the academic content. The purpose of a quality article is to inform and provide outcomes, allowing readers to apply it as they see fit.
Each of the questions should be clearly addressed or answered in the article. If you are interested in learning whether a rehab program would be appropriate for high school athletes, the studies you review should have been conducted with people in the same age range. This is called generalizability - that is, can the results of a study be applied to another population based on similarities in the populations?
Authors should clearly tell you how they selected participants and why, as well as any reasons participants were excluded from participation. Did the author report issues with the data that could not be explained, and was the number of participants at the start and end of the study reported? If fewer completed the study than initiated it, was there an explanation for the dropout rate? If an author was researching the use of latex bandages to decrease swelling after ankle sprains and a volunteer participant had an allergy to latex, that person would need to leave the study. In this case, an author should report that 40 subjects volunteered to participate and were initially taking part in the study, and one was removed because of an exclusionary criterion. The author should not say that 39 people agreed to participate.
Similarly, if a new test (index test) is being assessed, it should be compared to an appropriate gold standard reference. If a researcher is testing a new axillary thermometer to establish whether it accurately assesses core body temperature, it should be compared to rectal thermometer measures. Comparing a new axillary thermometer to oral thermometers or others that are not scientifically accepted as the gold standard could introduce significant error, which would make the study results inappropriate to use in clinical practice.
Also important when appraising an article is assessing its methods section. Is it detailed enough that you could replicate it, if desired? You should be able to fully understand what was done during the study when reading the methods section.
The prevention of bias is also important during a study. If a participant was assessed using both an indexed and a referenced test, were the results interpreted separately? In other words, did the person completing the assessments know the results of the comparative assessment? A person who is aware of the other outcomes when completing an assessment creates a potential for biased reporting and a loss of objectivity, which is important in research.
Learn more about Evidence-Based Practice in Athletic Training.More Excerpts From Evidence Based Practice in Athletic Training
Get the latest insights with regular newsletters, plus periodic product information and special insider offers.