Nursing research

Transcription

Good afternoon. I'm Sharon Hewner and I am Assistant Professor in the University of Buffalo School of Nursing. I've been asked to talk today about the difference between evidence-based practice (called EBP) and quality improvement (or QI). I'll add one other confusing term to the discussion, nursing research (which we'll label NR). It's easy to get nursing research (NR), evidence-based practice (EBP) and quality improvement (QI) confused because all three use data collection and analysis to answer important questions in nursing and health care. NR, EBP, and QI have been described as the 3 legs of a 3-legged stool that support clinical decision-making practice and nursing practice. This diagram shows how the 3 legs support nursing practice and is taken from a 2006 article by Christine Hedges called "Research, evidence-based practice, and quality improvement. The 3-legged stool." Let's start by defining the three terms. Nursing research is a systematic inquiry that uses disciplined methods to answer questions or solve problems. The goal is to generate new knowledge. Typically, research is based on theory and uses rigorous methods. Results are disseminated widely in peer-review publications and individuals must give consent to participate in the process. Quality improvement is a cyclical process designed to evaluate work flow and work processes. Data is collected as in research, but usually there are no theoretical underpinnings. QI does not seek to generate new knowledge but rather to improve patient outcomes. The results and data are typically used internally and reported only within the institution. Furthermore, usually Institutional Review Board approval is not required because the goal is to apply knowledge to existing care processes. Evidence-based practice is a problem solving approach to clinical practice that integrates a systematic search for, and critical appraisal of the most relevant research to answer a practical clinical question, one's own clinical expertise, and patient preferences and values. EBP is based on a rigorous, systematic review of existing research, particularly randomized control trials or RCT. So, how do these 3 legs support nursing practice? Let's look at an example from clinical practice. Suppose we are interested in preventing ventilator-associated pneumonia or VAP. In our hospital, monitoring the incidence of ventilator-associated pneumonia shows that rates have gone up - this is QI. To address the issue, critical care staff conduct extensive review of the literature and national guidelines, that is, they use EBP. After completing the review, staff may have identified research gaps that might be addressed using Nursing Research. And finally, staff may develop an approach to implement best practice and evaluate the results of the intervention. This is QI, again. When you are asked how the policy on prevention of ventilator-associated pneumonia was developed, you would say that all three legs of the stool are needed to answer the clinical question. QI is used to monitor practice and evaluate patient outcomes. EBP is focused on systematically reviewing the evidence and Nursing Research conducts scientific studies when the evidence is lacking. Now, I would like to dig just a little bit deeper into the processes involved in quality improvement. The Institute of Healthcare Improvement has developed an Open School for Health Professionals which has 6 six modules on quality improvement. The first lesson of module 3 presents the Model for Improvement which asks 3 questions. First, what are we trying to accomplish? This question includes defining the problem and efforts to understand complexity. Knowing that the goal is to reduce rates of VAP helps to establish the scope of the problem to be addressed. The second question asks, how will we know that the change is an improvement? There are a number of tasks involved which include identification of outcomes, establishing baseline measures and developing sampling strategies. Finally what change can we make that will result in improvement? This is where we apply the evidence to identify an approach that has worked in other situations. In this process, data can accelerate QI. People who are using data for improvement typically are working on the front lines of health care. They are not researchers who live in an ivory tower. They're people who want patients to get better care-fast. For example, if you want to develop a plan to improve hand-washing compliance, collecting data is going to help you get where you want to go. Specifically, data will help you understand current performance such as "Our staff members are washing their hands before only 40% of all encounters with patients." This is usually referred to as developing a baseline. Secondly, data will help you come up with ideas to improve the process. You might say that "It seems like compliance is lower among the weekend staff - let's target our changes at them." The third approach is that data will help you test changes to see if they lead to improvement. For example, "After we installed hand sanitizer dispensers on the 3 West medical unit, compliance rose to 95%." Data will help you ensure those improvements are being maintained. So that "Four months after we implemented this change, compliance remains above 95%." But measuring for improvement is quite different from measuring for research. When you're working on an improvement project you don't pick a hypothesis and stick with it until the bitter end. In improvement work, if you see that change is not leading to improvement, you adjust your hypothesis, and through a series of small tests of change or Plan, Do, Study, Act (PDSA) cycles, you refine the change until you work out all the kinks. Now this next table is adapted from Solberg, Mosser, and McDonald's 1997 article entitled, "The three faces of performance measurement: improvement, accountability, and research. The table looks at the type of study looking at comparing improvement and research. Please review table and compare Improvement and Research Please review table and compare Improvement and Research In improvement work, we accept consistent biases while in research, it's designed specifically to eliminate biases. The sample size in improvement projects is just enough data. In research you always want to get enough data to demonstrate statistical significance. Please review table and compare Improvement and Research Compared to research where the hypotheses are fixed prior to the initiation of the study. Please review table and compare Improvement and Research Please review table and compare Improvement and Research Please review table and compare Improvement and Research Here are some more tips to help you measure for improvement from the Institute for Healthcare Improvement (however, these won't help you if you're doing a big research study for the National Institutes of Health.) The first is, if a change isn't working, adjust it. That's not a weakness of the study design; it's an essential element of it Accept that various types of bias can exist in the data, but rather than attempt to eliminate that bias by controlling for it, try to keep it consistent. Improvement research doesn't take place in isolated controlled conditions; typically it happens in real work settings. Again, this isn't a problem; it's an essential source of learning about how to make the change work reliably; for every patient, every time, in every setting. Whenever possible, develop measures from data that someone is already collecting. That will make your project a lot simpler without compromising on quality. When it's necessary to develop new measures, try to make the data relatively easy to obtain. So in conclusion, patients receive the best care possible when nurses are dedicated to the continuous review and revision of their practice. This includes using quality improvement, evidence-based practice and research. Thank you.