Welcome back to Nursing2009 Critical Care's new regular column, Research Rounds. In this column, critical care nurses who meet regularly will discuss key research principles exemplified by a recently published article. Our goal is to demystify the research process and give you tips so that reading and applying research can be your best tool for success.
In this issue, our research discussion focuses on the format of a typical research article. Understanding the format will help you "speed read" a research article by initially focusing on the results to decide if your practice could benefit from the research. Should you find results you're interested in replicating, you can read the entire article carefully, focusing on the section on limitations, to determine whether the research has substance, was conducted well, and delivered trustworthy results.
In this article, we'll discuss the format used to publish research by examining an article that compares the benefits of two different kinds of education and counseling for older adults with heart failure.1 Research articles, including this one, follow a particular format. Let's consider the topics covered, one by one.
Introduction: Making a case
The opening paragraphs capture our interest by describing the problem and discussing its significance. In this case, the researchers discuss implementing an intervention designed to improve quality of life for the older adult with heart failure (HF). The authors must emphasize the key issue being researched-in this case, how to deliver affordable, personalized behavioral support that reaches most patients. Their search of the literature revealed that no previous studies had been conducted about the feasibility and effectiveness of a Web-based HF management program. By finding little or no research in a certain area, they had identified a gap in the literature.
Finally, the last sentences of the introductory section usually state the purpose of the research. Although space is limited, exemplary articles at least mention the model used to direct the research. These researchers use a known model called Participatory Action Research. To explore a phenomenon, a model can be used to direct and more fully explain the phenomena. In other words, the model is a "lens" through which we view our research.
For example, if our "lens" says that baking a consistently perfect cake is a function of time, oven temperature, and batter, we should be able to bake a cake to perfection each time we use the same time, oven temperature, and batter. The model would involve variables of time, oven temperature, and batter to yield a perfect cake. When we observe that baking the perfect cake each time involves more than these three variables, because the variables are the same but the results differ, we found our research question. When our "lens" added the key variable of adjusting for altitude, the model variables become time, oven temperature, batter, and altitude to yield a perfect cake. Using a known model that has previously explained phenomena strengthens the research process.
Research methods: Aiding replication
The research methods section describes the procedure followed. This section includes the topics of research design, ethics, criteria for selection of participants, and the reliability and validity of the instruments to measure the variables. Institutions appoint an Institutional Review Board (IRB) whose key responsibility is to protect the rights of the patients involved in the research process. Ethics includes having a review by the IRB, using the informed consent process, and obtaining permission to contact the patients. Researchers should explain their rationale for selecting the criteria that make patients eligible or ineligible to participate in the research and clearly delineate these criteria. Well-established tools with acceptable levels of reliability and validity must be used to measure the variables.
Novice researchers may naively entertain the idea of developing their own instruments. Instrument development is a complex, extensive process that should be piloted carefully over time and not undertaken lightly.
Because they're clinical studies, nursing studies aren't usually experimental research completed in a research lab under highly controlled conditions. Instead, they're usually quasi-experimental-group research, which means that researchers examine the effects of nursing interventions on nursing outcomes. The researchers in this study used two groups: one received Web-based intervention and the other received traditional HF care management.
The Web-based intervention included three modules: pathophysiology and self-care management, risk factor modification, and psychosocial and spiritual issues related to HF. Research subjects in this group could use an "E-mail the Expert" feature to ask their provider non-emergency care questions.
We were interested to see how the researchers addressed the older participants' access to Web-based intervention. We found that study participants didn't need internet access because they could access the intervention through their television.
The Control Attitude Scale, used in this research to measure the variable of perceived control, has high instrument reliability. The SF-12, used in this research to measure mental and physical health, is a well-known, reliable instrument.
The methods section should clearly explain how the research was conducted. A reader should be able to replicate the study and achieve the same results in another situation.
Results: Worthwhile for your practice?
The outcome of the research is presented in the text, with more details being provided in tables and figures. As in this article, the first table usually portrays demographics of the study participants. Tables, which are more succinct than text, give the reader a quick overview. For this study comparing two groups, the researchers can also demonstrate the differences and the approximate equivalence of the groups.
The second table demonstrates the mean values of the subject scored on three baseline and outcome variables. The significance of the change over time is measured. Two of three variables, mental health and perceived control, were statistically significant, as indicated by an asterisk.
Statistical significance means that the results are due to more than chance alone. The change for the control group is statistically significantly different than the change for the Web-based group for mental health and for perceived control. This Web-based intervention was thought to have enhanced mental health and perceived control after 3 months.
Figures help the reader to visualize the change in results from baseline to 3 months in the variables under observation. The first figure depicts the change in physical health. The second figure depicts the changes in perceived control over time. The third figure depicts the changes in mental health. Although these figures provide visual cues for the results, the reader must focus on the second table to know whether these changes indicate statistically significant differences.
The results presented in this article suggest that a Web-based education and counseling program can be implemented to achieve benefits in mental health and perceived control for older HF patients.
Discussion: Defining the limitations
Researchers are expected to comment on the study and the findings, including any limitations they identified. These researchers did an excellent job of discussing limitations of their study. Among other points, they discussed two biases that could have affected the results of the study. The first bias is called selection bias. The design of the study included participant volunteers from the HF clinic who were primarily white, thus preventing generalization to other races and cultures. The second bias is observational or learning biases. For example, the participants' mental health might have improved because they received a great deal of personal attention.
Building on your knowledge
Integrating what you learned in the last Research Rounds, you could next seek out review articles that summarize similar research to strengthen your confidence in the results. You'll be well on your way to improving practice for your patients.
REFERENCE