Immediate Action Research Methods: The quick and dirty on evidence and how to read a research paper.
Anecdotal is a dirty word. Very dirty, even more so when dealing with evidence-based medicine and life-saving issues. Understanding why anecdotal evidence that is sensationalized and massively marketed is an important building block to best practices in healthcare, whether it's happening with your family or in a professional healthcare setting. Additionally, being able to read, comprehend, and understand if a research article is valid and worthwhile is also a vastly underdeveloped and under-practiced skill.
Anecdotal, Logical, and Empirical Evidence
Anecdotal evidence is information that is collected non-scientifically. Many times they are told as a narrative via social or mass media and may contain large amounts of bias. There are a plethora of types of bias out there, but confirmation bias is perhaps one that largely dominates the tactical medical world… an example of this would be something along the lines of “well, I used it, and it saved my buddy’s life, so it works just as good as everything else.” This can lead to dogma, which has dominated fields like pre-hospital and tactical medical care.
Logical evidence more or less follows its namesake. The best example of this is to channel the Star Trek character Spock. When he seeks to make an argument, or defend an action he has chosen, he does so with a train of thought that makes sense and is support by facts that come from research and empirical evidence.
Empirical evidence is evidence that is supported by research that follows the Scientific Method. This evidence followed a defined model and research method that is reported with as little bias as possible, and more importantly, repeatable. For example, say a new medical device has become popular through marketing and social media response and even has one video supporting it. This really holds no water against other devices on the market that have been studied via the scientific method, and many times are found in peer reviewed studies. A great source for medical issues dealing tactical/operational medical care and tourniquets is the Journal of Special Operation Medicine.
How to read a research paper
Research papers can be intimidating to read by many, and used inappropriately by others. The first step after locating a paper you want to read is identifying if it is a primary source or a review article. The later means the article was written by the researchers, who generated the data being analyzed and presented. A review article may collect multiple primary articles and seek to analyze the data on a larger scale.
The first mistake many people make when attempting to use an article to defend a belief is reading the abstract, instead of the introduction. You cannot gain a real understanding of the article by just skipping to the abstract. Secondly, identifying the ‘big’ question is your next step. What are the research authors trying to answer with the data they collected. Thirdly, summarize what is going on: was the question answered, what data was collected, what issues with the research were identified, and what future research do the authors suggest needs to happen?
The fourth and fifth steps piggyback off each other. After you have figured out the broad question the authors are trying to answer, are there specific questions and answers the authors are trying to find? For example, you may have an article that looks at general fluid resuscitation of trauma patients but specifically looks at colloids versus crystalloid fluids. From this, the fifth step, you should identify the process in which the authors undertook to answer these specific questions.
Perhaps the most important part, the methods section, this is the how to why that is answered. There many, many semesters worth of graduate level education to be had about research methods. There is not really a quick and dirty way to learn about research methods, but the end result is, does this study make sense, can it be repeated, and is it done in a way that produces the data the authors say it generated?
The results section covers steps seven and eight. Looking at the results, what does the author report, and do they reject their hypothesis or prove it? How have they represented this proof? Via charts, graphs, equations etc? A largely important part of this is the sample size, commonly referred to as n= (whatever the number is). Larger sample sizes, many times, but not always lead to better results. The easy answer? Be more skeptical of studies with smaller, least diverse populations.
While it seems like a lot of work, it really is not. While it can take some practice to totally understand what is going on when you read the article, being able to interpret the information, and use it promote or refute an argument is a highly powerful tool, and worth at least having a basic understanding of the process.
Until next time, go crack open JSOM, and read some articles on tourniquets…