Respiratory Solutions

How to Critically Read a Scientific Paper

• By Phillip Dukes
• Jun 01, 2009

When working in the respiratory field, it's imperative to keep up with the latest research to make sure you are providing the best care for your patients. But how can you know if the research you're reading is relevant to your business practices? Critical analysis is crucial to understanding a scientific paper. Below are some steps to improving your critical analysis.

1. Before reading a paper, try to understand each figure, table and graph. The text will eventually help you to understand these graphics completely, but this step will put you in “question-asking mode.”

2. Read the paper from beginning to end, highlighting things that you don’t understand. Every scientific paper is a narrative of what the researchers thought and did. The formal structure of a modern paper is sometimes referred to as AIMRAD or Abstract, Introduction, Methods, Results, Analysis and Discussion.

The abstract (overall synopsis), introduction (reasons for the research), methods (how it was conducted and analyzed), results (what was found) and discussion (what the results mean) are the meat and potatoes of the paper and should lead you to whether or not the conclusions are valid.

The introductory sentence of a scientific paper should state the background of the research and should be followed by a brief review of the literature and an outline of the assumptions that will not be tested. Some authors may be listed repeatedly, which indicates that they have done a lot of work in this particular area. The hypotheses are what is tested in the research. Sound critical analysis requires the reader to distinguish between the assumptions and hypotheses.

The methods section gives the critical reader answers to questions about the design of the research: Who participated? How was it done? What was measured? Were the participants randomized? Did the participants try both the placebo and treatment, known as “crossover”? How were the results examined?

A good research paper will end with a discussion section, which should interpret the findings and discuss relevance. The authors should describe the limitations of the research.

3. Consider the statistics. Most research articles use statistics to form conclusions. Statisticians will use terms like mean, median and standard deviation.

Mean is a fancy word for average. It's the sum of all the values, divided by the number of subjects in the study. The median is the score or value that falls closest to the middle. Standard deviation describes how spread out the data is. These statistics are descriptive and intended to give you good information on what the subjects and the data look like. Are the means and medians similar? Are the findings very spread out?

The next level of statistics attempts to show differences between two or more groups or relationships between two or more things. When researchers test for statistical significance, they compare different sets of values — such as blood pressure before and after using CPAP — while taking into account how many people participated in the research, how dramatic the findings seem to be and what some of the characteristics were of the people compared.

Researchers then use mathematical formulas to calculate probability values. For the CPAP and blood pressure study, the probability value would inform us how likely it is that people in the study got better because the CPAP did its job or whether they got better simply because of chance or an unknown factor. If the probability value is low (less than 5 percent, 1 percent or even 1/10th of 1 percent), the conclusion is that the CPAP does work.

These probability values — called p values — represent percentages and are typically represented as p‹.01 or p‹.001. For example, p‹.01 means that there is less than 1 percent chance that blood pressure decreased because of chance alone. If probabilities are low, researchers describe them as statistically significant. These are keywords you should look for as you read. Remember, the lower the p value, the smaller the percentage, the greater the significance and the less likely that something happened just because of chance.

How significant is significant enough? It depends on what's being studied, the potential benefit or harm of the findings, and the author's standards. If the CPAP device, for example, is relatively inexpensive and safe, and high blood pressure isn't life threatening, the small chance that the CPAP is not as effective as indicated (a significance of p‹.05, or less than 1 in 20 chance that it works) may not be a big deal. If, instead, the researchers were studying the effects of cancer drugs that cause severe side effects, you want to be sure it's likely to work — the lower the p value, the better.

This article originally appeared in the Respiratory Management June 2009 issue of HME Business.