Snapshot: This document provides an overview of nursing research, a summary of the research process, including discussion of the popular “Iowa Model,” in addition to research terminology and a library of links.
Nursing Research: Overview & Scope
- Nursing research is a general term for the systemic investigation of the methods, goals, values and outcomes of nursing practice.
- As is true for other healthcare professionals, many nurses rely upon custom, habit or institutional norms to guide nursing practice. One of the central goals of nursing research is to subject such practice to rigorous, scientific inquiry.
- Basic research is a catch-all term for scientific research aimed at advancing fundamental knowledge without any immediate goal or practical use in mind. Most nursing research is more applied and goal-driven, although to the extent that nursing research (and research in other healthcare professions) depends upon a broad scientific understanding of the human body and disease processes, nursing research is indirectly dependent upon basic research, and benefits from this research.
- An example of basic research would be investigating the way cancer cells use signaling pathways to multiply. The research might or might not subsequently yield insights into the pathophysiology of cancer, and to new cancer treatments.
- Evidence-based practice (EBP) is a subset of nursing research. The goal of EBP is to examine accepted practice in the light of external clinical evidence to identify practice that is or is not supported by research using sound methodology. EBP therefore identifies the best evidence to guide nursing practice.
- An example of EBP would be encouraging early ambulation as tolerated for post-surgical patients because a randomized control trial demonstrated that this practice led to a statistically significant reduction in rates of pneumonia and shorter hospital stays.
- Quality improvement (QI) or performance improvement (PI) is subset of nursing research that is concerned with improving processes and outcomes at the institutional level. Among other goals, QI/PI aims to help organizations work in ways that are more safe, effective, timely and efficient. QI/PI is a process that encourages continuous, ongoing experimentation and improvement.
- An example of QI/PI would be formulating a set of clinically validated guidelines for reducing falls, such as mandating that RN’s complete a fall-risk assessment on admission, and then measuring and re-measuring over time the implementation and impacts of such a tool on fall rates within the organization.
- Nursing research – in all its various forms – comprises one ‘leg’ of the ‘three-legged stool’ of nursing education, nursing practice, and nursing research. Ideally, these three activities are interdependent: each is mutually dependent upon the other, and contributes to betterment of the profession.
The Research Process
Research is not simply a body of knowledge. It is a dynamic process with distinct steps and phases. Below is a broad overview of that process.
- Formulate the Question. The first step in any research project is to decide what will be investigated. The best research questions are clear, concise and focused. For example, the research question “what challenges do new nurses face?” is too broad, and would be very difficult to investigate. A better question might be, “what are the leading challenges to retaining bachelors-prepared nurses in their first year of practice in a major academic medical center?”
- Define the Purpose or Goal. Nursing research is an inherently applied discipline: its purpose is to improve the breadth and depth of nursing knowledge, and to thereby improve nursing practice. When you conduct research, keep in mind your intended audience, and what you hope to achieve with your research. If you cannot answer these questions, you should return to your research question and re-formulate it.
- Choose a Theoretical Framework & Research Design. While nursing research is typically ‘practice oriented,’ it is nonetheless important to be cognizant of how the themes and concepts in your research fit together, and to spell out how you will investigate your topic for your audience. Will you be investigating individuals’ feelings and narratives? Will you be measuring the statistical significance of a particular clinical intervention? Will you be synthesizing the results of previous studies, or conducting original research? Your answers to these questions will help determine the purpose, methods, assumptions and limits of your research.
- Perform a Literature Review. Research never exists in a vacuum, and any research project should survey existing research. A competent literature review, however, is not simply a dry recitation of past findings. Rather, it should be a dynamic summary and analysis of existing knowledge that points out the links between this past work and the new research being conducted. The literature review is a useful introduction for the research audience that both situates the author’s research in a broader context, and helps to justify the importance of and need for the research question the author investigates.
- Select the Population and Sample. Regardless of your research goals, framework or methodology, you will need to define who will be included in your study, and who will not. The population or sample you investigate needs to match your research question, and just as importantly, must be available for you to study. Populations may be defined by any number of variables, such as age, gender, clinical role/status, or professional setting.
- Perform Ethical Review & Obtain Approval. An essential part of the research process is to ensure that your research topic is ethically sound, and to obtain clearances from your organization or institution to conduct the research. This concern is always important, but perhaps especially so for nurses, who often deal with private and highly sensitive medical data, access to which may be restricted by HIPAA and other regulations. You will probably need to present your research proposal to an Institutional Review Board (IRB), or other similar institution for approval. Plan ahead, because review and approval can be a lengthy process, and the panel may require that you make revisions or otherwise alter your research design.
- Collect & Analyze Data. Once you know your research design and methodology, you’ll need to gather and interpret your data. This data may be qualitative or quantitative, or a mix of both. Before you do your collection, make sure you have the knowledge and means to make sense of your data. You may need to use statistical analysis for quantitative research, or determine your methodology for analyzing subjective feelings, values and narratives for qualitative research. Again, allow time: data collection and analysis is often challenging and time-consuming.
- Organize, Write & Edit. Your research question, methodology, and population sample should help you structure your write-up of the project. One way to structure your paper is to think in terms of broad categories: introduction, methods, results, discussion, and conclusions and recommendations. Using headings and sub-headings to ‘break up’ your research is often helpful to readers. Also be sure to acknowledge the limitations of your research and areas for further inquiry.
- Compose Citations, Sources & Bibliography. Every research study must clearly acknowledge sources and properly cite them. To do so is not only important for reasons of intellectual integrity, but also because readers should be able to reference your sources should they choose to do further reading and investigation. Follow your organization or institution’s accepted format; one commonly used format in nursing research is the American Psychological Association (APA) style.
- Communicate Findings. This final stage in the research process is a crucial one. After all, research that never finds its way to its intended audience is inert. Communicating your findings often includes publication; it may also include speaking about and presenting your research to professional audiences and to the public. However this communication occurs, be open to feedback. The questions and debates your research inspires may form the basis for your next investigation, or for research conducted by others in conversation with your own.
The Iowa Model
- The Iowa Model of Evidence-Based Practice to Promote Quality Care is a practice-driven research framework that has become a standard in professional nursing and nursing research.
- The Iowa Model begins by prompting staff nurses to identify clinical research problems, triggered either by identification of a question or challenge in nursing practice (a “problem-focused trigger”), or by encountering new knowledge from existing research (a “knowledge-focused trigger.”)
- The topic should be aligned with an organizational priority. If it is not, the Model prompts the nurse researcher to consider a different “trigger,” or basis for the research.
- Once a topic has been selected, the next steps in the Model dictate forming a team, and then assembling, synthesizing, reviewing and critiquing the available research on the topic.
- If the team discovers that there is sufficient evidence in the research to support the proposed practice change, the team can proceed with ‘piloting,’ or testing, the change to determine its usefulness, feasibility and impact.
- Should the pilot project prove successful or encouraging, the practice change can then be ‘rolled out’ and gradually integrated within the organization on a larger scale.
- The team must continue to communicate and collaborate throughout this process to select evidence-based practice driven goals, collect data, monitor processes and outcomes, and modify the practice change as appropriate.
- The team is encouraged to communicate internally to implement the change, and externally with other organizations and stakeholders to expand nursing knowledge and to encourage the wider adoption of evidence-based practice driven changes in nursing.
- If, on the other hand, the team discovers that there is not sufficient evidence in the research to support the proposed practice change, the team has two primary options.
- First, the team could choose to pursue the practice change on the basis of weaker forms of evidence, such as case reports and expert opinion. Alternately, the team could also choose to conduct its own research, and once further research has been completed, then return to the question of whether there exists a sufficient research basis for piloting and potentially implementing the proposed practice change.
- The Iowa Model integrates theory and practice to help professional nurses discover, formulate and advocate for evidence-based practice changes. It also possesses the great advantage of never leaving practitioners without a ‘next step.’ No matter what the professional nurse and his or her team encounter, the Model has clear recommendations and guidelines for a course of action.
Research Concepts & Terminology
- Research Hypothesis. The supposition or proposed explanation a researcher wishes to investigate or test. The hypothesis should propose a relationship between two or more variables.
- Null hypothesis. The supposition that there is no relationship between the variables or phenomena a researcher is investigating.
- Control group. The group of individuals in an experiment who are not subjected to the experimental intervention. The control group serves as a benchmark for measuring the effects of the experimental intervention on the treatment group.
- Independent vs. dependent variable. In experimental research, the independent variable is the “input” variable that is believed to influence or affect the dependent variable, or “output” variable. The independent variable is therefore unchanged by experimental manipulation, whereas the dependent variable is changed by such manipulation.
- Inductive vs. deductive reasoning. Inductive reasoning is the “bottom up” process of reasoning from specific examples or observations to formulate general principles. In contrast, deductive reasoning is a “top down” process of reasoning from the basis of one or more general principles to formulate more specific hypotheses, predictions, or explanations of phenomena.
- Qualitative vs. quantitative analysis. Qualitative analysis is a way of interpreting non-numerical data to draw out the themes, meanings and relationships between phenomena. In contrast, quantitative analysis is a way of interpreting numerical data, often using statistical analysis, to assess the significance and magnitude of the causal relationships among variables.
- Observational studies. An observational study is one in which the researcher observes the behaviors of a group of subjects without intervening in the group in order to draw inferences or conclusions. Observational studies stand in contrast with experiments, such as randomized control trials, which have treatment and control groups. Observational studies include panel studies, cohort studies and case-control studies.
- Cohort study. A cohort study is a type of forward-looking observational study in which a cohort (group) of individuals is followed over time. The cohort is usually investigated at regular intervals, and the characteristics or behaviors of the cohort are compared with the general population from which the cohort is drawn.
- Case-control study. A case-control study is a type of backward-looking observational study in which a group of individuals who share a common characteristic such as a disease or medical condition (“the cases”) are compared with one or more groups who do not possess the characteristic (“the controls”) in order to identify what caused the condition in “the cases”.
- Statistical vs. practical significance. In research, a statistically significant result is one that is not the result of chance or sampling error, thereby allowing the researcher to reject the null hypothesis. However, a statistically significant result may lack practical significance if the size or impact of the finding is not large or clinically significant.
- Type I vs. Type II error. In statistics, a Type I Error or “false positive result,” is defined as incorrectly rejecting a true null hypothesis. A Type II Error or “false negative result” is defined as failing to reject a false null hypothesis. Consequently, a Type I Error will lead a researcher to falsely conclude that a meaningful relationship or effect exists, whereas a Type II Error will lead a researcher to fail to detect a meaningful relationship or effect.
- Hierarchy of Evidence. A hierarchy of evidence is a way of grading the quality and relative authority of various types of research studies. Systematic reviews and meta-analyses are typically at the top of the hierarchy, followed by randomized control trials, cohort, case control and cross sectional studies; case studies, expert opinion and anecdotal observations are typically at the bottom. Many researchers believe clinical guidelines should be based on the highest possible level of evidence.
- Meta-Analysis. A meta-analysis is “review of reviews,” or an analysis of multiple research studies in order to draw out the studies’ most well-supported findings and conclusions. Meta-analysis is employed in many systemic reviews, and according to many researchers occupies the highest level of evidence in the hierarchy of evidence.
- Double-Blind Experiment. A double-blind experiment is one in which neither the person conducting the experiment (the researcher) or the subjects of the experiment know information about the experiment that could lead to conscious or unconscious bias. For example, if a researcher was evaluating whether a drug was more effective than a placebo, in a double-blind experiment neither the researcher nor the test subjects would know which group of patients received the drug under investigation, and which received the placebo.
- External vs. internal validity. In research, a study with findings that can be easily generalized to the general population is said to possess external validity. A study with high internal validity is one in which confounding variables have been successfully eliminated, and the causal relationship under investigation has been clearly established by the researcher. These forms of validity each exist on a continuum, and a study may be high in both kinds of validity, low in both, or high in one but not in the other.
- Reliability vs. validity. Reliability is concept in statistics and psychometrics that refers to the overall consistency of a given type or method of measurement. There are several different kinds or reliability, such as inter-method reliability (the degree to which different methods for measuring a given variable are consistent), inter-rater reliability (the degree to which the measurements or ratings made by different individuals of a given variable are consistent), and test-retest reliability (the degree to which a given test yields the same results when repeated over time). In contrast, validity measures the degree to which a given measurement is measuring what a researcher intends to measure.
Links & Resources
- National Institute of Nursing Research (NINR): http://www.ninr.nih.gov
- NINR funds and promotes professional nursing research, and trains and educates current and future nurse research professionals.
- The website has a repository of video and audio resources for nurse researchers, provides news of interest to the nursing and broader medical research community, and offers summaries of the latest funded NINR research. Check out the website’s Grant Development and Management Resources page, which includes valuable information, links and resources for both research grant applicants and funded nurse researchers.
- The organization also offers both “intramural” (on-site) and “extramural” (off-site) research programs. Research encompasses a wide range of professional nursing and practice concerns, from health disparities to disease prevention to end of life care. NINR staff can also assist in each stage of the grant application process – from preparation, to review, to funding and post-grant management.
- ANA Nursing World: http://www.nursingworld.org/EspeciallyForYou/Nurse-Researchers
- The American Nursing Association (ANA) offers a Research Toolkit that offers a valuable overview of and introduction to nursing research and evidence based practice. The Toolkit offers links to numerous research resources, and access to a repository of reviews of research articles. You have register as a member on the site to get access.
- Essential Nursing Resources (ENR) provides a compilation of print and online resources of interest to nurse researchers. The Table of Contents clearly breaks down resources by category, and each resource is helpfully flagged as online or print, and fee- or no-free required for access.
- The National Database of Nursing Quality Indicators (NDNQI), part of the ANA’s National Center for Nursing Quality (NCNQ), is a national nursing quality measurement program that offers hospitals nursing unit-level performance data, benchmarked against state, regional and national averages. Such data is often valuable to nurse researchers. NDNQI currently has over 1,500 participating U.S. hospitals.
- Association for Healthcare Research & Quality (AHRQ): http://www.ahrq.gov/professionals/clinicians-providers/resources/nursing/index.html
- The AHRQ supports research to subject clinical practice to critical, evidence-based standards, and to improve health outcomes. AHRQ is making a concerted effort to encourage contributions to nurse researchers to this broader mission of health research.
- Nursing research funded by AHRQ is summarized on the AHRQ Nursing Research web page. AHRQ provides extramural (off-site) funding for nursing-related health research, and follows the same peer-review process as NIH (the National Institutes of Health).
- The agency offers a Research Activities Online Newsletter that summarizes study findings of interest to professional nurses and nurse researchers.