One of the most exciting trends in nursing is the proliferation of clinical nursing research and the subsequent development of practice policies based on research findings. The development of procedure policies involves conducting a critical evaluation of research published in a given area, such as nasogastric feeding, and using the research to develop and evaluate a protocol for a particular clinical area.
Most nurses are comfortable with developing procedures, training staff, and evaluating outcomes of care with quality assessment monitors. However, some nurses have never taken a research course, and some nurses work in areas where there are few nurses with advanced educational preparation or physicians that are interested in research. To these nurses, the process of evaluating research articles to determine whether to integrate new interventions into practice can appear overwhelming. One does not have to be an expert, though, to critique research and to determine its use in a clinical setting.
It is helpful to form a group of nurses to evaluate articles together, discuss findings, and build enthusiasm. Select an area of practice on which to focus, such as skin care, pain control, wound healing, or quality assessment/ quality improvement. The area should be relevant to the clinical setting and a body of research knowledge in the area should exist. For example, a topic found in the Journal of Gerontological Nursing would be useful.
Once the selection is made, contact your nearest professional library. It could be a university, community college, or hospital library. Most have a librarian who can help you search for research articles. Many centers have computerized systems that can search through several years of journals in a matter of minutes, locating pertinent information and often article abstracts. Look for studies and articles that review bodies of research literature. Don't expect to find an overwhelming number of articles, though. Most subject searches generate relatively few research studies - especially in nursing. Once the articles are located, scan the references for original sources or additional research studies noted in the articles' literature reviews. Make copies of the articles for everyone to read before meeting.
Now you are ready to evaluate the studies (Figure).
Examine the Journal
Is it peer reviewed? In the front of refereed journals (such as the Journal of Gerontological Nursing) there is usually an editorial board listed. These are the people who read submitted manuscripts and advise the editor on the quality of the article. Note who comprises the editorial board. Are they nurses? If so, what are their credentials? In your opinion, would these people be able to identify quality research?
Look at the Authors' Backgrounds
In each journal there will be a short statement identifying the authors, their positions, and places of employment. Note their credentials. Are the authors nurses or allied health care professionals? Do they work for an institution that supports clinical research, such as a university or a hospital? Do any of the authors have research credentials such as a PhD, DNS, or EdD?
Critically Read the Study
* What is this study about? What was the purpose of the study and the research question asked? What did the authors base the study on? Did they have a theory? Did they cite other studies or literature that convinced you the study was important?
* What were the study methods? Did the author describe how the study was conducted? Who were the subjects? Were they people or animals? Changing practice based on animal studies is not recommended without trials in humans. If the subjects were people, were they similar to the population you serve? For example, studies of benefits of exercise in college-age males will not generalize to a longterm care population.
* How many subjects were there? In general, the more subjects, the stronger the study. A good starting number may be 30 to 50. It is important not to develop a procedure or policy based on one study with small sample size (less than ten subjects). Look for several studies on the same topic using similar subjects resulting in comparable findings. Practice changes should not be based on studies with preliminary findings only. The purpose of preliminary studies may be to introduce new ideas and encourage further research in the area.
Four Steps for Evaluating a Research Article
* If an intervention or medication was tested, was there a control group who did not get the treatment? If not, were the subjects evaluated before and after receiving the treatment using the same measures as during the intervention?
* Where was the study conducted? Will the setting generalize to your clinical area?
* Did the study answer the research question? Does the study describe how it was carried out? Pay attention to how data were collected. Was there a questionnaire? Did the authors measure something? Did the questionnaire or measurement seem appropriate? Did the authors report that it had been used before or tested prior to the study? Questionnaires and measures should be studied before they are used to collect data to see if they are valid and reliable, meaning they collect the information in a consistent and accurate manner.
* What do the statistics tell you? Don't be intimidated by the word "statistics." Look at the numbers. Do the columns in the tables add up? Are numbers given in percentages? It is often better to rely on sample scores (numbers) rather than percentages. For example, knowing that 60% of a study sample preferred chocolate milk to regular milk may sound conclusive until you find that there were only five people in the sample.
* Did the author find statistical significance in the findings? Statistical significance will be reported by the author when it occurs. Be cautious of changing practice based on opinion surveys or anecdotal reports. Be equally cautious of people selling products based on their research. While some may be valid, careful critique is usually warranted.
* What are the implications for clinical practice? What does the author identify as implications for clinical practice? In your opinion, are the implications or recommendations for practice appropriate, or is there a huge leap from the data to the implications? For example, does the author recommend giving chocolate milk to every resident in a long-term care setting based on the preferences of 3 out of 5 subjects?
* Does the author discuss limitations of the study? Because many of our subjects are human, nursing research is not easy to conduct. Every study has flaws and limitations. The author should identify some in the article such as having a small sample, a short time span for data collection, or studying only men. Does the author identify needs for future research?
Seek Expert Advice
Don't expect to be an expert. Many universities have journal clubs to teach faculty and students how to evaluate research. If there is a gerontology or nursing journal club in your area, ask to attend.
Evaluating research, like any skill, takes practice. Don't hesitate to contact nursing faculty or another research-trained health professional (usually someone with a master's degree or higher) to help you. If you still have questions about the study, contact the researcher. This is commonly done and is welcomed by many nurse researchers. Writing for publication is never easy - and one reward is hearing from enthusiastic readers.
WHY SHOULD I?
Research evaluation can sharpen critical thinking and enhance nursing practice and patient outcomes. Moreover, it can add an exciting new dimension to patient care, to quality improvement programming, and to your selfidentification as a professional. You do not have to be a genius to read research reports critically if you have the human and material resources available.