Nurse Anesthesia Research: Science of an Orderly, Purposeful, and Systematic Nature

Chapter 4


Nurse Anesthesia Research


Science of an Orderly, Purposeful, and Systematic Nature



The certified registered nurse anesthetist (CRNA) brings a wealth of knowledge to the clinical arena. Although this knowledge comes from a variety of disciplines, including physiology, pharmacology, physics, nursing, medicine, and psychology, it should be appreciated that research and critical thinking first and foremost make this knowledge possible. Evidence-based practice greatly enhances credibility within the clinical setting.


Research represents a rational approach to the making of practice choices among initially plausible alternatives and provides direction and a means for validating these choices. Whether selecting one intravenous opioid over another or choosing one particular pediatric induction technique instead of another, CRNAs rely on research to provide a solid foundation for clinical decision making, thereby avoiding fads and inferior alternatives.


The impact of research on the day-to-day activities of the CRNA has become an especially relevant topic. Before the mid-1970s the vast majority of nurses functioned without much consideration of research or publication of their ideas. In the late 1970s we experienced a period of punctuated evolution. Major driving forces behind this evolution included movement into a graduate educational framework, a more sophisticated appreciation of the scientific underpinnings of our specialty, recognition of the importance of evidence-based practice (EBP; see the discussion of this topic later in this chapter), national attention to issues of patient outcome and patient safety, and a growing self-awareness of nurse anesthetists not only as providers of excellent clinical care but also as active participants as scholars in the field.


Because CRNAs primarily function with a practice-oriented perspective, the recommendations of Brown et al1 seem especially relevant. These scholars suggested that four characteristics of research are essential for the development of a scientific knowledge base for a discipline such as nurse anesthesia. First, research should be actively conducted by the members of the discipline. Second, research should be focused on clinical problems encountered by members of the discipline. Third, the approach to these problems must be grounded in a conceptual framework—that is, it must be scientifically based, emphasizing selection, arrangement, and clarification of existing relationships. And finally, the methods used in studying the problems must be fundamentally sound.



Ways of Knowing


The term research can be broadly defined as the application of a systematic approach to the study of a problem or question. However, we do not know all the things we claim to know on the basis of systematic inquiry. For example, tradition and custom are important sources of human knowledge. Those who live in the United States are raised in a democratic society and are taught that democracy is the best and most advanced form of government. This is a powerful and efficient route for communication of knowledge because it excuses individuals from initiating an independent effort to come to grips with the concept of democracy. In the absence of evaluation for validity, however, such a route may lead to blind acceptance.


Another source of our knowledge is authority. We know something to be true because an authoritative person such as a parent, educator, clergyman, physician, or teacher tells us it is true. Yet, despite the fact that authorities are fallible, the knowledge they pass on often remains unchallenged. Should we not ask the basis for what we are being told?


Personal experience (the trial-and-error method) represents a powerful source of knowledge. We make observations (e.g., that placing a hand on a hot stove causes a burn) and on their basis make predictions (e.g., that a stove may be hot) and future behavioral decisions (e.g., to avoid touching a stove). However, a risk remains: Not only are certain events perceived differently by different people, but one person’s experience may be too narrow to serve as the basis for the development of a reasonable and unbiased understanding of a given phenomenon. Although this mechanism is a practical way of knowing, it is highly fallible and represents a coarse and inefficient way to gain knowledge.


Logical reasoning is yet another way of knowing. The reasoning method has two components: inductive reasoning and deductive reasoning. Inductive reasoning results in generalizations that are derived from specific observations. Consider the following line of reasoning using a character in many action and adventure movies, James Bond, for example. We observe that James Bond is mortal; we observe that a number of other people are mortal as well; on this basis, we conclude that all people are mortal. Deductive reasoning is the development of specific predictions from generalities. In this case, we see the following line of reasoning: we know that all men are mortal; we know that James Bond is a man; therefore we conclude that James Bond is mortal. Both methods are useful, but the former offers no mechanism for evaluation or self-correction, and the latter is not in itself a source of new information.


Perhaps the most advanced way of knowing is reflected in the scientific method. Although it too is fallible, the scientific method is more reliable and valid than other methods. It provides for self-evaluation with a system of checks and balances that minimizes bias and faulty reasoning. In essence, it is a systematic approach to solving problems and enhancing our understanding of phenomena. It has, at its foundation, the gathering and interpretation of information without prejudice.



The Nature of Research


Research is by definition a dynamic phenomenon. Whether it is directed purely at the acquisition of knowledge for knowledge’s sake (basic research) or at the specific solution of problems (applied research), it is a process that can be conceptualized in terms of at least four characteristics.


First, research can assume many different forms. Second, research must be valid, both internally and externally (Box 4-1). Internal validity is necessary but not sufficient for ensuring external validity. Third, research must be reliable. Reliability refers to the extent to which data collection, analysis, and interpretation are consistent and to which the research can be replicated. Fourth, research must be systematic. The elements of a systematic approach include the identification of the problem or problems, the gathering and critical review of relevant information, the collection of data in a highly orchestrated manner, an analysis of the data appropriate to the problem or problems faced, and the development of conclusions within the study’s framework.



BOX 4-1   Research Scenario: Internal Versus External Validity




External Validity


The extent to which the results can be generalized; this issue relates to the question “To whom can the results be applied?”


For example, 35 obese men who are nonsurgical volunteers are anesthetized with a standard dose of a new induction drug. The clinical half-life of the drug is determined with plasma drug sampling and brain wave activity monitoring. The researcher concludes that future patients receiving the standard dose of the new drug will experience a clinical half-life of 11 minutes.




Science is not a routine, cut-and-dried process. Rather, scientific knowledge emerges from an enterprise that is intensely human; as a consequence, it is subject to the full spectrum of human strengths and limitations. The scientific discovery and understanding that attend participation in research and its results can be professionally exhilarating and satisfying.



The Eight Critical Stages in the Research Process


Research accords several personal freedoms to those who engage in it: the freedom to pursue those opportunities in which one is interested, the freedom to exchange ideas with other interested colleagues, and the freedom to be a deconstructionist—that is, one who challenges existing knowledge. Yet, despite these freedoms, research must be logical, must progress in an orderly manner, and ultimately must be grounded within the framework of the scientific method. If research is a way of searching for truths, uncovering solutions to problems, and generating principles that result in theories, we must come to understand the process of research.


The research process can be described in many different ways. For purposes of simplicity, this process is defined as consisting of the following eight distinct stages:




Stage 1: Identification of the Problem


The selection and formulation of the problem constitute an essential first step in the research process. The researcher decides the general subject of the investigation, guided principally by personal experience and by inductions and deductions based on existing sources of knowledge. The researcher makes the general subject manageable by narrowing of the focus of the problem. The following criteria must be met at this phase of the research process:



We constantly encounter problems and situations that can be studied. At clinical anesthesia conferences, one might hear remarks such as the following:



A study could emerge from each of these situations, built on ideas, hunches, or curiosity. A problem that lends itself to research often materializes from personal observations and in the sharing of ideas and experiences among those who are familiar with the phenomenon in question.


Once identified, the problem should be stated in terms that clarify the subject and restrict the scope of the study. Defining the terms involved in the problem statement also is critical, as demonstrated in Box 4-2.



BOX 4-2   Research Scenario: Stating the Problem





The wording of the problem statement sets the stage for the type of study design used. Each step in the research process subsequently influences later steps, and this should be kept in mind at all times. A mistake made early inevitably creates difficulties at some later stage in the process. The novice researcher may be surprised to find that this first stage in the research process often consumes a large portion of the total time invested in the research effort. Yet the time is well spent, because research should not commence until a problem has been identified and formulated in a thoughtful and useful manner.




Stage 2: Review of the Relevant Knowledge and Literature


Once the problem has been identified, information is needed for putting the problem into proper context so that the research can proceed effectively. A well-conducted literature review provides the researcher with the following:



The knowledge that influences the problem originates from three general sources: personal files and experience, personal contacts with experts, and the library and Internet. Both manual indexes and computerized databases should provide the researcher with immediate and full access to the world’s published literature. Additional literature searches may be required at different times throughout the research process.




Stage 3: Formulation of the Hypothesis or Research Question


In its most elemental form, a hypothesis is either a proposition of the solution to a problem or a stated relationship among variables. It establishes and defines the independent variable (the variable that is to be manipulated or is presumed to influence the outcome) and the dependent variable (the outcome that is dependent on the independent variable). The hypothesis is declarative in nature and assumes one of the following three forms:



Research questions are generally reserved for investigations that are descriptive or exploratory in nature or for when the relationships among the variables are unclear. A research question might be more appropriate than a hypothesis in a study that proposes to determine the beliefs of anesthesia providers who interact with patients under specific circumstances. For example, consider the following research question: What are the attitudes of CRNAs in the northeastern United States who care for patients with acquired immunodeficiency syndrome (AIDS)?




Stage 4: Development of an Approach for Testing the Hypothesis


After the research idea has taken shape in the form of a formal hypothesis or research question, a plan of attack is developed. The research proposal represents the stage at which the ideas of the project crystallize into a substantive form. The proposal includes the following:



A research proposal is a useful and efficient way for the researcher to determine the completeness of the plan and is usually required if the researcher is to obtain departmental or institutional approval or is applying for financial support.



Research Methods


The research method is the way the truth of a phenomenon is coaxed from the world in which it resides and is freed of the biases of the human condition. A variety of research methods are at our disposal, and researchers are not inflexibly wedded to any particular approach. Researchers do not follow a single scientific method but rather use a body of methods that are amenable to their fields of study.


Some of the methods available are highly recognizable, permanent components of the researcher’s armamentarium, whereas others have evolved not only with respect to time but also in response to the specific needs of a particular problem or discipline. The research method can be influenced by the way a researcher views a problem. For example, a researcher can test a hypothesis, search for a correlation, ask “why” or “how” questions, or probe a phenomenon on the basis of “what would happen if” suppositions.


The researcher can view the method on the basis of the fundamental task that it will accomplish. For example, two broad categories into which research efforts can be divided are basic research and applied research. Basic research adds to the existing body of knowledge and may not have immediate, practical use. Applied research is oriented toward solving an immediate, specific, and practical problem.


The research method can be characterized in terms of its temporal relationship to the problem. A retrospective study is the process of surveying the past; the thing in which we are interested has already occurred, and we are simply looking to see what did occur. In contrast, a prospective study looks forward to see what will happen in a given situation; here, the collection of data proceeds forward in time.


It is important to understand several terms fundamental to the research process. As mentioned previously, the dependent variable is the object of the study, or the variable that is being measured. The independent variable is the one that affects the dependent variable and is presumed to cause or influence it. Another way of looking at this relationship is that variables that are a consequence of or are dependent on antecedent variables are considered dependent variables.


Another set of variables consists of control variables, also known as organismic, background, or attribute variables. Control variables are not actively manipulated by the researcher, but because they might influence the relationships under study, they must be controlled, held constant, or randomized so that their effects are neutralized, canceled out, or at least considered by the researcher (Box 4-3).



BOX 4-3   Research Scenario: Understanding the Types of Variables




The term blinding refers to the process of controlling for obvious and occult bias arising from subjects’ or researchers’ reactions to what is going on. In a single-blind design, the patients are unaware of which treatment or manipulation is actually being given to the subjects. In the double-blind design, neither the researcher nor the subject is aware of which treatment or manipulation the subject is receiving. Whereas randomization attempts to equalize the groups at the start of the study, blinding equalizes the groups by controlling for psychological biases that might arise apart from any effect of the treatment. Many factors influence the decision to use a single-blind or a double-blind design. For example, in some situations, it may not be feasible to disguise a particular treatment or intervention.


Operationalization is the process of making the characteristics inherent in a given variable, condition, or process familiar or clear to others. If researchers do not operationalize the terms, phrases, and manipulations in the study, the net effect could be an ambiguous study. For example, in a study examining the effects of epidural anesthesia in critically ill patients, it would be essential to operationalize the terms effects and critically ill patients. Similarly, in a study comparing the quality of inhalation induction with isoflurane and sevoflurane in pediatric patients, it is essential that the researcher operationalize the terms inhalation induction and quality. Operationalization of terms clearly designates performable and observable acts or procedures in such a way that they can be replicated immutably.



Classifying Research on the Basis of Methodology


Although different authors use a variety of classification schemes, the following example provides a simple way for the researcher to select and classify a design. This scheme attends to the study’s purpose and scope and to the nature of the problem at hand. Table 4-1 offers a simplified approach to classifying research design.



Quasi-experimental research differs from experimental research in that it is missing one or more of the key elements required for the experimental design. Either a control group or a randomization procedure may be absent from the design. For example, at an institution, outpatients may routinely receive ondansetron from a particular practitioner, whereas they routinely do not receive the drug from another practitioner. A prospective trial in which both practitioners use a standard anesthetic technique could be initiated. For example, isoflurane, an opioid, and cisatracurium could be administered; this would allow the two practitioners to use or not use ondansetron as they normally would. Outcome, measured in terms of the incidence of nausea and vomiting in the first 6 postoperative hours, is quantified, and the groups are compared. Although randomization is not achieved, a study that may not otherwise have been possible because of the inflexibility of the clinicians involved is successfully accomplished. Quasi-experiments, by yielding to one or more of the rigid criteria of the experimental design, offer an attractive alternative in certain circumstances.



Qualitative Research: An Alternative Paradigm


Up to this point, the traditional approach to a problem has been characterized by deductive reasoning, objectivity, manipulation, and control. An alternative approach involves a group of methods characterized by inductive reasoning, subjectivity, exploration, and process orientation. These methods fall under the rubric of qualitative research techniques.


Qualitative techniques include philosophic inquiry, historiography, phenomenology, grounded theory, and ethnography. Generally speaking, qualitative research refers to systematic modes of inquiry directed principally at observing, describing, analyzing, interpreting, and understanding the patterns, themes, qualities, and meanings of specific contextual phenomena. Qualitative research seeks to gain insight by discovering the meanings associated with a given phenomenon and exploring the depth, richness, and complexity inherent in it.


For example, exploring how male and female CRNAs differ in the manner in which they deal with parental and child separation when a child is readied for induction of anesthesia might best be achieved through the use of a qualitative design. The actual experiences might be observed or videotaped. Those involved—anesthetists, parents, and children—might be interviewed immediately and at some time after the procedure. This study would be artificially constrained and disjointed if it were conducted in any setting other than the original one or if too many controls were brought to bear on the experiment.


The qualitative paradigm seems especially appropriate when the researcher does not want to artificially distance a study from its contextual richness or when there is not enough information available on a particular subject for the adequate development of sound and testable hypotheses. The treatise on qualitative approaches by Marshall and Rossman2 is recommended to interested readers.



Sampling


Under most circumstances, studying everyone who might be affected by a particular study is impractical, if not impossible. For example, if we want to know how effective intravenous nitroglycerin is in minimizing the rise in blood pressure associated with laryngoscopy in hypertensive patients, we cannot realistically study all hypertensive patients who undergo laryngoscopy. Rather, we would hope to find a smaller group of subjects who are representative of the relevant population at large. By accessing certain information in the sample, we can credibly make inferences or generalizations regarding the population at large.


Similarly, if we want to know how often anesthesia machines in small community hospitals receive preventive maintenance, we cannot visit all the community hospitals in the nation. Instead, we might randomly select a number of hospitals in a number of different states, visit those locations, and inspect the maintenance records. By studying this representative sample, we can make some reasonable and safe generalizations regarding the phenomenon of preventive maintenance at large.


Consider the anecdote about the four blind people who encountered an elephant during one of their daily walks. Each person felt a different part of the elephant. When asked to describe what they had encountered, the first person replied, “a tree trunk” (the elephant’s leg). The second reported feeling “a large snake” (the elephant’s trunk). The third reported that it was “most definitely a wall” (the elephant’s torso). The last person reported that it was “a large, frayed rope” (the elephant’s tail). This analogy illustrates that a few discrete sampling points may not be adequate for describing a complex phenomenon. Not only is a random sample best, it also should be large enough and sample a sufficient number of points in the population that a truly representative perspective is gained.


Different sampling techniques can be used, depending on the research design used. In a true random sample (also known as probability sampling), all members of the population at large have a similar chance of being included in the study. This is rarely the case in clinical research, in which we are confined to dealing with those individuals who present themselves. In this situation, the sample is called a convenience sample. When a convenience sample is used in an experimental study, it is important to ensure that the subjects selected for the study are at least randomized when assigned to treatments or groups. In the ideal situation, the researcher aims for both random selection (from the population at large) and random assignment (to the different groups in the study).


For example, in a study designed to quantify the rate of arterial desaturation in pediatric patients who are transported to the postanesthesia care unit with and without supplemental oxygen, the researcher is limited to those patients who are undergoing surgery. It is difficult to obtain a sample from the pediatric population at large and subject them to anesthesia and surgery. Rather, a convenience sample of patients who are having an operation is used. However, the researcher should randomly assign the study participants to one of the two treatment groups—those who receive supplemental oxygen or those who do not receive supplemental oxygen.


Obtaining a random sample, especially in clinical research, is often a complicated process. Most important is the realization that the concept of randomness is essential to minimizing human biases associated with both selection and assignment.



Instrumentation and Measurement


Two important concepts essential to measurement are validity and reliability. Instrument validity is the degree to which an instrument, such as a blood pressure cuff or a personality inventory, measures what one believes it is measuring. Instrument reliability refers to the degree of consistency with which an instrument measures whatever it is measuring—that is, whether the same result is obtained on repeated trials.


Validity and reliability are often easily established for measures of certain physiologic phenomena but may be troublesome in behavioral or psychological evaluations. Imagine trying to determine reliability and validity for a thermometer. Contrast this to trying to establish validity and reliability for a psychological tool that professes to measure a CRNA’s attitude toward euthanasia; obviously, the latter is a much more difficult undertaking. Although a measure must be reliable to be valid, it can be reliable without being valid. For example, a skin temperature probe might reliably (consistently) measure temperature even in a variety of extreme settings, although it would not be viewed as a valid indicator of core temperature. Both reliability and validity are discussed in degrees rather than in “all-or-nothing” terms.


Many published instruments have reliability and validity testing reported. When choosing an instrument for a study, it is critical to consider whether the instrument’s reliability and validity have been established. For example, if an instrument measures evoked responses in the esophagus as an indicator of depth of anesthesia, it must be determined whether the reliability and validity of the instrument have been established under the conditions of the anesthetic protocol being used in the proposed study. Coefficients of reliability and validity are presented on a scale of 0 to 1, with 1 being perfect.


Occasionally the researcher may encounter no reasonable measures to use for a study. For example, instruments for measuring such phenomena as arterial oxygen tension, end-tidal anesthetic concentration, and opioid metabolic by-products are well established. A researcher may need to develop a totally new instrument (questionnaire) to determine perceptions regarding the propriety of a given manufacturer’s high-pressure promotional campaigns for newly released pharmaceutical products. In developing such a tool, it is helpful to have an expert in the discipline look over the instrument and provide feedback to ensure that the instrument is appropriate.


Researchers have a variety of instruments for measuring phenomena. These include the following:




Levels of Measurement

In designing a study, the researcher must decide how to measure a phenomenon such as anxiety level, blood pressure, attitude toward health care, or rate of complications. There are four levels or degrees of measurement: nominal, ordinal, interval, and ratio. The type of data measured determines the kind of statistical analysis that can be done. Table 4-2 characterizes the four levels of measurement.



TABLE 4-2


Characteristics of the Four Categories of Measurement























Only gold members can continue reading. Log In or Register to continue

Related posts:

Outpatient Anesthesia Obesity and Anesthesia Practice Renal Anatomy, Physiology, Pathophysiology, and Anesthesia Management Positioning for Anesthesia and Surgery Musculoskeletal System Anatomy, Physiology, Pathophysiology, and Anesthesia Management Regional Anesthesia: Upper and Lower Extremity Blocks
Tags:
May 31, 2016 | Posted by in ANESTHESIA | Comments Off on Nurse Anesthesia Research: Science of an Orderly, Purposeful, and Systematic Nature

Full access? Get Clinical Tree

 Get Clinical Tree app for offline access
Category Characteristics Examples
Nominal Identifies Male or female
    Diagnosis
Ordinal Identifies American Society of Anesthesiologists (ASA) class
  Orders Order of race finish