Future posts are now at http://discoveringyourinnerscientist.wordpress.com/

The DISCUSSION Section of a Research Article:

What does the Results Section Mean?

After reading the last few blog posts you should now be able to recognize a published research report as having these key sections:
Title
Abstract
Introduction
Methods 
Discussion

Key Point: Now we turn to DISCUSSION section  where the researcher tells you what the Results section means.   Discussion gives "the investigator's explanations of some of their findings and their opinions of what they think their study means for practice & future research" (p. 22, Davies & Logan, 2008).

Let’s see how this works by looking at an example of Results & then Discussion in Apfelbaum et al’s 2013 study of “Postoperative pain experience” published in Anesthesia and analgesia.

In the RESULTS section Apfelbaum et al’s statistical analysis results were:

Approximately two thirds of patients reported that a health care professional talked with them before surgery about how their pain would be treated….Overall, nurses were more likely than other health care professionals to educate patients about pain and pain management. Among surgical outpatients, surgeons were as likely as nurses to provide patient pain education, but nurses were more likely to provide this service for surgical inpatients. After surgery, two thirds of patients reported being asked by a health care professional about their pain, most frequently by a nurse….75% of patients believed that it was necessary to experience some pain after surgery, and 8% of patients had postponed surgery because they were worried about the possibility of experiencing pain (pp. 537-8)

You can see that the Results above tells only results of statistical tests.  In the DISCUSSION section Apfelbaum and co-authors explained what the Results mean for RNs & providers.

We were surprised to find that more than half of the patients surveyed were concerned about experiencing pain after surgery and that this caused some of them even to postpone surgery. Although most patients claimed to receive preoperative education on postoperative pain management, our findings suggest that a patient’s real concern is not adequately addressed.  Despite the increased focus on pain management over the last several years and the development of formal standards and guidelines for the management of acute pain, a significant number of patients continue to experience unacceptable levels of pain after surgery and after discharge. This fact is alarming, considering the trend toward ambulatory surgery and shorter hospital stays. (p. 539)

Critical thinking:  

FIRST read the following Results of a study by Fayh et al (2013) that was an experiment to test “the effects of 5 % weight loss, through diet only or diet plus exercise, on lipid profile, inflammation and endothelial function in obese individuals.”  THEN write your own Discussion section explaining what these results mean. 

Results: Thirteen individuals dropped out before completing the weight loss intervention. The median time required for reduction of 5 % of initial body weight was 79.7 days for the Diet group and 65.9 days for the Diet + Exercise group ( P = 0.16). In both Diet( n = 18) and Diet + Exercise ( n = 17), total cholesterol (−15.8 ± 4.8 and −10.5 ± 4.9 mg/dL, respectively), triglycerides (−33.8 ± 10.0 and −39.4 ± 10.3 mg/dL, respectively) and hs-CRP (−1.35 ± 0.41 and −0.45 ± 0.43 mg/L, respectively) decreased significantly, and in a similar response. (p. 1443)

FINALLY (if you like) you can then compare your Discussion to their Discussion section by getting a copy of their article from the library: Fayh, A. et al (2013). Effects of 5% weight loss through diet or diet plus exercise on cardiovascular parameters of obese: A randomized trial. European Journal of Nutrition, 52(5) 1443-50. doi http://dx.doi.org.libproxy.csun.edu/10.1007/s00394-012-0450-1 

“I want RESULTS!”

(Analysis of Data to Answer the Hypothesis or Question)

The “Results” section of a research article tells you both 1) the characteristics of those who participated in the study (e.g., their gender, ethnicity, education, & so on); & 2) the data analysis used to answer the research question or hypothesis.   The Results section follows the methods section that is described in preceding blogs.

“Results” is where the author reports analysis of numbers data using statistics or analysis of word data by identifying common themes.  Don’t be afraid to read this section; & don’t let your eyes glaze over.   All this comes with practice.  Here are a few basics to get started.*

1st          The researcher has collected data (or measurements) about something in numbers (e.g., inches or test scores) or words (e.g., subjects’ descriptions of experiences) or data in both numbers and words.  The researcher will analyze numbers data using statistical tests and will analyze word data for recurring themes and ideas.  

2nd        The characteristics of the participants in the study will tell you whether the participants are similar to or different from those to whom you want to apply the results.

3rd         In statistical analysis when you see that a result is p<.05 this means that there is a 95% chance the result is right and a 5% chance it is wrong.  When you see p<.01, it means that there is a 99% chance the result is right and a 1% chance the result is wrong.

4th         In statistics the researcher will analyze number data to do at least one of these:

a.       Describe something (for example, What are RNs’ self-care practices);

b.      Find out whether two things are related to each other (for example, Is maternal age related to numbers of birth defects); or 

c.       Identify whether one thing is causing another (for example, Does ZMapp vaccine cause those with Ebola virus to get well).  

5th         Not surprisingly, statistical analysis that describes something is called descriptive statistics (4^th.a above). Examples are percents & averages.  In contrast, statistical analyses done to find relationships or cause and effect are called inferential statistics (4^th b.c. above).  Examples are correlation coefficients or t-tests.

6th          In word analysis, the researcher will be able only to describe something (e.g., what do RNs experience when returning to school).

*[For more on basic statistics see Halfens, R.J.G., & Meijers, J.M.M. (2013). Back to basics: An introduction to statistics. Journal of Wound Care, 22(5): 248-51.]

CRITICAL THINKING

1.       Is the following example, did Zerwekh et al do word analysis or statistical analysis?  Did they describe, correlate, or explain cause and effect? EXAMPLE: In Zerwekh et al (2002) study, analysis showed that barriers to pain management came from 5 sources: “within the patient, within the physician, within the family, within the nurse and within the healthcare organization” (p.85).  [Hint: see 1^st & 6^th above]

2.       In the following example, were Smith et al (2010) focusing on identifying relationships or cause and effect?  How might you describe the likelihood that they were right, based on the “p” levels?  EXAMPLE: “Better pain management was associated with increased emotional well-being (t = 2.11, p = 0.03). Number of hospitalizations was marginally associated with increased emotional well-being (t = 1.91, p < 0.06).” (p.83)