6. Includes information from professional sources and required reading to augment personal reflections/positions.

Read chapters 2, 5, and 7 in your LoBiondo-Wood & Haber book

Chapter 2

Research questions, hypotheses, and clinical questions

Judith Haber

At the beginning of this chapter, you will learn about research questions and hypotheses from the perspective of a researcher, which, in the second part of this chapter, will help you generate your own clinical questions that you will use to guide the development of evidence-based practice projects. From a clinician’s perspective, you must understand the research question and hypothesis as it aligns with the rest of a study. As a practicing nurse, developing clinical questions (see Chapters 19, 20, and 21) is the first step of the

evidence-based practice process for quality improvement programs like those that decrease risk for development of pressure ulcers.

When nurses ask questions such as, “Why are things done this way?” “I wonder what would happen if . . . ?” “What characteristics are associated with . . . ?” or “What is the effect of ____ on patient outcomes?”, they are often well on their way to developing a research question or hypothesis. Research questions are usually generated by situations that emerge from practice, leading nurses to wonder about the effectiveness of one intervention versus another for a specific patient population.

The research question or hypothesis is a key preliminary step in the research process. The research question tests a measureable relationship to be examined in a research study. The hypothesis predicts the outcome of a study.

Hypotheses can be considered intelligent hunches, guesses, or predictions that provide researchers with direction for the research design and the collection, analysis, and interpretation of data. Hypotheses are a vehicle for testing the validity of the theoretical framework assumptions and provide a bridge between theory (a set of interrelated concepts, definitions, and propositions) and the real world (see Chapter 4).

For a clinician making an evidence-informed decision about a patient care issue, a clinical question, such as whether chlorhexidine or povidone-iodine is more effective in preventing central line catheter infections, would guide the nurse in searching and retrieving the best available evidence. This evidence, combined with clinical expertise and patient preferences, would provide an answer on which to base the most effective decision about patient care for this population.

Often the research questions or hypotheses appear at the beginning of a research article, but may be embedded in the purpose, aims, goals, or even the results section of the research report. This chapter provides you with a working knowledge of quantitative research questions and hypotheses. It also highlights the importance of clinical questions and how to develop them.

Developing and refining a research question: Study perspective

A researcher spends a great deal of time refining a research idea into a testable research question. Research questions or topics are not pulled from thin air. In Table 2.1, you will see that research questions can indicate that practical experience, critical appraisal of the scientific literature, or interest in an untested theory forms the basis for the development of a research idea. The research question should reflect a refinement of the researcher’s initial thinking. The evaluator of a research study should be able to identify that the researcher has:

• Defined a specific question area

• Reviewed the relevant literature

• Examined the question’s potential significance to nursing

• Pragmatically examined the feasibility of studying the research question

Defining the research question

Brainstorming with faculty or colleagues may provide valuable feedback that helps the researcher focus on a specific research question area. Example: ➤ Suppose a researcher told a colleague that her area of interest was health disparities about the effectiveness of peer coaching or case management in improving health outcomes with challenging patient populations such as those who are homeless. The colleague may have asked, “What is it about the topic that specifically interests you?” This conversation may have initiated a chain of thought that resulted in a decision to explore the effectiveness of a nursing case management and peer coaching intervention on hepatitis A and B (HAV and HBV) vaccine completion rates among homeless men recently released on parole (Nyamathi et al., 2015).

Beginning the literature review

The literature review should reveal a relevant collection of studies and systematic reviews that have been critically examined. Concluding sections in such articles (i.e., the recommendations and implications for practice) often identify remaining gaps in the literature, the need for replication, or the need for additional knowledge about a particular research focus (see Chapter 3). In the previous example, the researcher may have conducted a preliminary review of books and journals for theories and research studies on factors apparently critical to vaccine completion rates for preventable health problems like HAV and HBV, as well as risk factors contributing to the disproportionate impact of HAV and HBV on the homeless, such as risky sexual activity, drug use, substandard living conditions, and older age. These factors, called variables, should be potentially relevant, of interest, and measurable.

 EVIDENCE-BASED PRACTICE TIP

The answers to questions generated by qualitative data reflect evidence that may provide the first insights about a phenomenon that has not been previously studied. Other variables, called demographic variables, such as race, ethnicity, gender, age, education, and physical and mental health status, are also suggested as essential to consider. Example: ➤ Despite the availability of the HAV and HBV vaccines, there has been a low completion rate for the three-dose core of the accelerated vaccine series, particularly following release from prison. This information can then be used to further define the research question and continue the search of the literature to identify effective intervention strategies reported in other studies with similar high-risk populations (e.g., homeless) that could be applied to this population. Example: ➤ One study documented the effectiveness of a nurse case management program in improving vaccine completion rates in a group of homeless adults, but no studies were found about the effectiveness of peer coaching. At this point, the researcher could write the tentative research question: “What is the effectiveness of peer coaching and nursing case management on completion of an HAV and HBV vaccine series among homeless men on parole?” You can envision the interrelatedness of the initial definition of the question area, the literature review, and the refined research question.

Examining significance

When considering a research question, it is crucial that the researcher examine the question’s potential significance for nursing. This is sometimes referred to as the “so what” question, because the research question should have the potential to contribute to and extend the scientific body of nursing knowledge. Guidelines for selecting research questions should meet the following criteria: • Patients, nurses, the medical community in general, and society will potentially benefit from the knowledge derived from the study. • Results will be applicable for nursing practice, education, or administration. • Findings will provide support or lack of support for untested theoretical concepts. • Findings will extend or challenge existing knowledge by filling a gap or clarifying a conflict in the literature. • Findings will potentially provide evidence that supports developing, retaining, or revising nursing practices or policies. If the research question has not met any of these criteria, the researcher is wise to extensively revise the question or discard it. Example: ➤ In the previously cited research question, the significance of the question includes the following facts: • HAV and HBV are vaccine preventable. • Viral hepatitis disproportionately impacts the homeless. • Despite its availability, vaccine completion rates are low among high-risk and incarcerated populations. • Accelerated vaccine programs have shown success in RCT studies. • The use of nurse case management programs in accelerated vaccine programs also provides evidence of effectiveness. • Little is known about vaccine completion among ex-offender populations on parole using varying intensities of nurse case management and peer coaches. • This study sought to fill a gap in the related literature by assessing whether seronegative parolees randomized to one of three intervention conditions were more likely to complete the vaccine series as well as to identify predictors of HAV/HBV vaccine completion.

The fully developed research question

When a researcher finalizes a research question, the following characteristics should be evident: • It clearly identifies the variables under consideration. • It specifies the population being studied. • It implies the possibility of empirical testing. Because each element is crucial to developing a satisfactory research question, the criteria will be discussed in greater detail. These elements can often be found in the introduction of the published article; they are not always stated in an explicit manner.

Variables

Researchers call the properties that they study “variables.” Such properties take on different values. Thus a variable, as the name suggests, is something that varies. Properties that differ from each other, such as age, weight, height, religion, and ethnicity, are examples of variables. Researchers attempt to understand how and why differences in one variable relate to differences in another variable. Example: ➤ A researcher may be concerned about the variable of pneumonia in postoperative patients on ventilators in critical care units. It is a variable because not all critically ill postoperative patients on ventilators have pneumonia. A researcher may also be interested in what other factors can be linked to ventilator-acquired pneumonia (VAP). There is clinical evidence to suggest that elevation of the head of the bed and frequent oral hygiene are associated with decreasing risk for VAP. You can see that these factors are also variables that need to be considered in relation to the development of VAP in postoperative patients. When speaking of variables, the researcher is essentially asking, “Is X related to Y? What is the effect of X on Y? How are X1 and X2 related to Y?” The researcher is asking a question about the relationship between one or more independent variables and a dependent variable. (Note: In cases in which multiple independent or dependent variables are present, subscripts are used to indicate the number of variables under consideration.) An independent variable, usually symbolized by X, is the variable that has the presumed effect on the dependent variable. In experimental research studies, the researcher manipulates the independent variable. In nonexperimental research, the independent variable is not manipulated and is assumed to have occurred naturally before or during the study. The dependent variable, represented by Y, varies with a change in the independent variable. The dependent variable is not manipulated. It is observed and assumed to vary with changes in the independent variable. Predictions are made from the independent variable to the dependent variable. It is the dependent variable that the researcher is interested in understanding, explaining, or predicting. Example: ➤ It might be assumed that the perception of pain intensity (the dependent variable) will vary in relation to a person’s gender (the independent variable). In this case, we are trying to explain the perception of pain intensity in relation to gender (i.e., male or female). Although variability in the dependent variable is assumed to depend on changes in the independent variable, this does not imply that there is a causal relationship between X and Y, or that changes in variable X cause variable Y to change.

Although one independent variable and one dependent variable are used in the examples, there is no restriction on the number of variables that can be included in a research question. Research questions that include more than one independent or dependent variable may be broken down into subquestions that are more concise. Finally, it should be noted that variables are not inherently independent or dependent. A variable that is classified as independent in one study may be considered dependent in another study. Example: ➤ A nurse may review an article about depression that identifies depression in adolescents as predictive of risk for suicide. In this case, depression is the independent variable. When another article about the effectiveness of antidepressant medication alone or in combination with cognitive behavioral therapy (CBT) in decreasing depression in adolescents is considered, change in depression is the dependent variable. Whether a variable is independent or dependent is a function of the role it plays in a particular study.

Population

The population is a well-defined set that has certain characteristics and is either clearly identified or implied in the research question. Example: ➤ In a retrospective cohort study studying the number of ED visits and hospitalizations in two different transition care programs, a research question may ask, “What is the differential effectiveness of nurse-led or physician-led intensive home visiting program providing transition care to patients with complex chronic conditions or receiving palliative care (Morrison, Palumbo, & Rambur, 2016)? Does a relationship exist between type of transition care model (nurse-led focused on chronic disease self-management or physician-led focused on palliative care and managing complex chronic conditions) and the number of ED visits and rehospitalizations 120 days pre- and post-transitional care interventions?” This question suggests that the population includes community-residing adults with complex chronic conditions or receiving palliative care who participated in either a nurse or physician-led transitional care program.

Testability

The research question must imply that it is testable, measurable by either qualitative or quantitative methods. Example: ➤ The research question “Should postoperative patients control how much pain medication they receive?” is stated incorrectly for a variety of reasons. One reason is that it is not testable; it represents a value statement rather than a research question. A scientific research question must propose a measurable relationship between an independent and a dependent variable. Many interesting and important clinical questions are not valid research questions because they are not amenable to testing.

Study purpose, aims, or objectives

The purpose of the study encompasses the aims or objectives the investigator hopes to achieve with the research. These three terms are synonymous. The researcher selects verbs to use in the purpose statement that suggest the planned approach to be used when studying the research question as well as the level of evidence to be obtained through the study findings. Verbs such as discover, explore, or describe suggest an investigation of an infrequently researched topic that might appropriately be guided by research questions rather than hypotheses. In contrast, verb statements indicating that the purpose is to test the effectiveness of an intervention or compare two alternative nursing strategies suggest a hypothesis-testing study for which there is an established knowledge base of the topic. Remember that when the purpose of a study is to test the effectiveness of an intervention or compare the effectiveness of two or more interventions, the level of evidence is likely to have more strength and rigor than a study whose purpose is to explore or describe phenomena. Box 2.1 provides examples of purpose, aims, and objectives.

Study purpose, aims, or objectives

The purpose of the study encompasses the aims or objectives the investigator hopes to achieve with the research. These three terms are synonymous. The researcher selects verbs to use in the purpose statement that suggest the planned approach to be used when studying the research question as well as the level of evidence to be obtained through the study findings. Verbs such as discover, explore, or describe suggest an investigation of an infrequently researched topic that might appropriately be guided by research questions rather than hypotheses. In contrast, verb statements indicating that the purpose is to test the effectiveness of an intervention or compare two alternative nursing strategies suggest a hypothesis-testing study for which there is an established knowledge base of the topic. Remember that when the purpose of a study is to test the effectiveness of an intervention or compare the effectiveness of two or more interventions, the level of evidence is likely to have more strength and rigor than a study whose purpose is to explore or describe phenomena. Box 2.1 provides examples of purpose, aims, and objectives.

Relationship statement

The first characteristic of a hypothesis is that it is a declarative statement that identifies the predicted relationship between two or more variables: the independent variable (X) and a dependent variable (Y). The direction of the predicted relationship is also specified in this statement. Phrases such as greater than, less than, positively, negatively, or difference in suggest the directionality that is proposed in the hypothesis. The following is an example of a directional hypothesis: “Nurse staff members’ perceptions of transformational leadership among their nurse leaders (independent variable) is that it is negatively associated with nurse staff burnout (dependent variable)” (Lewis & Cunningham, 2016). The dependent and independent variables are explicitly identified, and the relational aspect of the prediction in the hypothesis is contained in the phrase “negatively associated with.” The nature of the relationship, either causal or associative, is also implied by the hypothesis. A causal relationship is one in which the researcher can predict that the independent variable (X) causes a change in the dependent variable (Y). In research, it is rare that one is in a firm enough position to take a definitive stand about a cause-and-effect relationship. Example: ➤ A researcher might hypothesize selected determinants of the decision-making process, specifically expectation, socio-demographic factors, and decisional conflict would predict post decision satisfaction and regret about their choice of treatment for breast cancer in Chinese-American women (Lee & Knobf, 2015). It would be difficult for a researcher to predict a cause-and-effect relationship, however, because of the multiple intervening variables (e.g., values, culture, role, support from others, personal resources, language literacy) that might also influence the subject’s decision making about treatment for their breast cancer diagnosis. Variables are more commonly related in noncausal ways; that is, the variables are systematically related but in an associative way. This means that the variables change in relation to each other. Example: ➤ There is strong evidence that asbestos exposure is related to lung cancer. It is tempting to state that there is a causal relationship between asbestos exposure and lung cancer. Do not overlook the fact, however, that not all of those who have been exposed to asbestos will have lung cancer, and not all of those who have lung cancer have had asbestos exposure. Consequently, it would be scientifically unsound to take a position advocating the presence of a causal relationship between these two variables. Rather, one can say only that there is an associative relationship between the variables of asbestos exposure and lung cancer, a relationship in which there is a strong systematic association between the two phenomena.

Testability

The second characteristic of a hypothesis is its testability. This means that the variables of the study must lend themselves to observation, measurement, and analysis. The hypothesis is either supported or not supported after the data have been collected and analyzed. The predicted outcome proposed by the hypothesis will or will not be congruent with the actual outcome when the hypothesis is tested.

Relationship statement

The first characteristic of a hypothesis is that it is a declarative statement that identifies the predicted relationship between two or more variables: the independent variable (X) and a dependent variable (Y). The direction of the predicted relationship is also specified in this statement. Phrases such as greater than, less than, positively, negatively, or difference in suggest the directionality that is proposed in the hypothesis. The following is an example of a directional hypothesis: “Nurse staff members’ perceptions of transformational leadership among their nurse leaders (independent variable) is that it is negatively associated with nurse staff burnout (dependent variable)” (Lewis & Cunningham, 2016). The dependent and independent variables are explicitly identified, and the relational aspect of the prediction in the hypothesis is contained in the phrase “negatively associated with.” The nature of the relationship, either causal or associative, is also implied by the hypothesis. A causal relationship is one in which the researcher can predict that the independent variable (X) causes a change in the dependent variable (Y). In research, it is rare that one is in a firm enough position to take a definitive stand about a cause-and-effect relationship. Example: ➤ A researcher might hypothesize selected determinants of the decision-making process, specifically expectation, socio-demographic factors, and decisional conflict would predict postdecision satisfaction and regret about their choice of treatment for breast cancer in Chinese-American women (Lee & Knobf, 2015). It would be difficult for a researcher to predict a cause-and-effect relationship, however, because of the multiple intervening variables (e.g., values, culture, role, support from others, personal resources, language literacy) that might also influence the subject’s decision making about treatment for their breast cancer diagnosis. Variables are more commonly related in noncausal ways; that is, the variables are systematically related but in an associative way. This means that the variables change in relation to each other. Example: ➤ There is strong evidence that asbestos exposure is related to lung cancer. It is tempting to state that there is a causal relationship between asbestos exposure and lung cancer. Do not overlook the fact, however, that not all of those who have been exposed to asbestos will have lung cancer, and not all of those who have lung cancer have had asbestos exposure. Consequently, it would be scientifically unsound to take a position advocating the presence of a causal relationship between these two variables. Rather, one can say only that there is an associative relationship between the variables of asbestos exposure and lung cancer, a relationship in which there is a strong systematic association between the two phenomena.

Testability

The second characteristic of a hypothesis is its testability. This means that the variables of the study must lend themselves to observation, measurement, and analysis. The hypothesis is either supported or not supported after the data have been collected and analyzed. The predicted outcome proposed by the hypothesis will or will not be congruent with the actual outcome when the hypothesis is tested.

Theory base

The third characteristic is that the hypothesis is consistent with an existing body of theory and research findings. Whether a hypothesis is arrived at on the basis of a review of the literature or a clinical observation, it must be based on a sound scientific rationale. You should be able to identify the flow of ideas from the research idea to the literature review, to the theoretical framework, and through the research question(s) or hypotheses. Example: ➤ Nyamathi and colleagues (2015) (see Appendix A) investigated the effectiveness of a nursing case management intervention in comparison to a peer coaching intervention based on the comprehensive health-seeking and coping paradigm developed by Nyamathi in 1989, adapted from a coping model by Lazarus and Folkman (1984), and the health-seeking and coping paradigm by Schlotfeldt (1981), which is a useful theoretical framework for case management, peer coaching interventions, and vaccine completion outcomes.

Statistical versus research hypotheses

You may observe that a hypothesis is further categorized as either a research or a statistical hypothesis. A research hypothesis, also known as a scientific hypothesis, consists of a statement about the expected relationship of the variables. A research hypothesis indicates what the outcome of the study is expected to be. A research hypothesis is also either directional or nondirectional. If the researcher obtains statistically significant findings for a research hypothesis, the hypothesis is supported. The examples in Table 2.4 represent research hypotheses. A statistical hypothesis, also known as a null hypothesis, states that there is no relationship between the independent and dependent variables. The examples in Table 2.5 illustrate statistical hypotheses. If, in the data analysis, a statistically significant relationship emerges between the variables at a specified level of significance, the null hypothesis is rejected. Rejection of the statistical hypothesis is equivalent to acceptance of the research hypothesis.

Directional versus nondirectional hypotheses

Hypotheses can be formulated directionally or nondirectionally. A directional hypothesis specifies the expected direction of the relationship between the independent and dependent variables. An example of a directional hypothesis is provided in a study by Parry and colleagues (2015) that investigated a novel noninvasive device to assess sympathetic nervous system functioning in patients with heart failure. The researchers hypothesized that participants with heart failure reduced ejection fraction (HFrEF), who have internal cardiac defibrillators or CRT pacemakers, will have a decrease in pre-ejection period (reflective of increased sympathetic nervous system activity) and decrease in left ventricular ejection time (reflective of an increased heart rate) with a postural change from sitting to standing. In contrast, a nondirectional hypothesis indicates the existence of a relationship between the variables, but does not specify the anticipated direction of the relationship. Example: ➤ Rattanawiboon and colleagues (2016) evaluated the effectiveness of fluoride mouthwash delivery methods, swish, spray, or swab application, in raising salivary fluoride in comparison to conventional fluoride mouthwash, but did not predict which form of fluoride delivery would be most effective. Nurses who are learning to critically appraise research studies should be aware that both the directional and the nondirectional forms of hypothesis statements are acceptable.

Relationship between the hypothesis and the research design

Regardless of whether the researcher uses a statistical or a research hypothesis, there is a suggested relationship between the hypothesis, the design of the study, and the level of evidence provided by the results of the study. The type of design, experimental or nonexperimental (see Chapters 9 and 10), will influence the wording of the hypothesis. Example: ➤ When an experimental design is used, you would expect to see hypotheses that reflect relationship statements, such as the following: • X1 is more effective than X2 on Y. • The effect of X1 on Y is greater than that of X2 on Y. • The incidence of Y will not differ in subjects receiving X1 and X2 treatments. • The incidence of Y will be greater in subjects after X1 than after X2.

 EVIDENCE-BASED PRACTICE TIP

Think about the relationship between the wording of the hypothesis, the type of research design suggested, and the level of evidence provided by the findings of a study using each kind of hypothesis. You may want to consider which type of hypothesis potentially will yield the strongest results applicable to practice.

Hypotheses reflecting experimental designs also test the effect of the experimental treatment (i.e., independent variable X) on the outcome (i.e., dependent variable Y). This suggests that the strength of the evidence provided by the results is Level II (experimental design) or Level III (quasi-experimental design). In contrast, hypotheses related to nonexperimental designs reflect associative relationship statements, such as the following: • X will be negatively related to Y. • There will be a positive relationship between X and Y. This suggests that the strength of the evidence provided by the results of a study that examined hypotheses with associative relationship statements would be at Level IV (nonexperimental design

Developing and refining a clinical question:

A consumer’s perspective Practicing nurses, as well as students, are challenged to keep their practice up to date by searching for, retrieving, and critiquing research articles that apply to practice issues that are encountered in their clinical setting (see Chapter 20). Practitioners strive to use the current best evidence from research when making clinical and health care decisions. As research consumers, you are not conducting research studies; however, your search for information from clinical practice is converted into focused, structured clinical questions that are the foundation of evidence-based practice and quality improvement projects. Clinical questions often arise from clinical situations for which there are no ready answers. You have probably had the experience of asking, “What is the most effective treatment for . . . ?” or “Why do we still do it this way?” Using similar criteria related to framing a research question, focused clinical questions form a basis for searching the literature to identify supporting evidence from research. Clinical questions have four components: • Population • Intervention • Comparison • Outcome These components, known as PICO, provide an effective format for helping nurses develop searchable clinical questions

The significance of the clinical question becomes obvious as research evidence from the literature is critically appraised. Research evidence is used together with clinical expertise and the patient’s perspective to confirm, develop, or revise nursing standards, protocols, and policies that are used to plan and implement patient care (Cullum, 2000; Sackett et al., 2000; Thompson et al., 2004). Issues or questions can arise from multiple clinical and managerial situations. Using the example of catheter acquired urinary tract infections (CAUTIs), a team of staff nurses working on a medical unit in an acute care setting were reviewing their unit’s quarterly quality improvement data and observed that the number of CAUTIs had increased by 25% over the past 3 months. The nursing staff reviewed the unit’s standard of care and noted that although nurses were able to discontinue an indwelling catheter, according to a set of criteria and without a physician order, catheters were remaining in place for what they thought was too long and potentially contributing to an increase in the prevalence of CAUTIs. To focus the nursing staff’s search of the literature, they developed the following question: Does the use of daily nurse-led catheter rounds in hospitalized older adults with indwelling urinary catheters lead to a decrease in CAUTIs? Sometimes it is helpful for nurses who develop clinical questions from a quality improvement perspective to consider three elements as they frame their focused question: (1) the situation, (2) the intervention, and (3) the outcome. • The situation is the patient or problem being addressed. This can be a single patient or a group of patients with a particular health problem (e.g., hospitalized adults with indwelling urinary catheters). • The intervention is the dimension of health care interest, and often asks whether a particular intervention is a useful treatment (e.g., daily nurse-led catheter rounds). • The outcome addresses the effect of the treatment (e.g., intervention) for this patient or patient population in terms of quality and cost (e.g., decreased CAUTIs). It essentially answers whether the intervention makes a difference for the patient population. The individual parts of the question are vital pieces of information to remember when it comes to searching for evidence in the literature.

Appraisal for evidence-based practice the research question and hypothesis

When you begin to critically appraise a research study, consider the care the researcher takes when developing the research question or hypothesis; it is often representative of the overall conceptualization and design of the study. In a quantitative research study, the remainder of a study revolves around answering the research question or testing the hypothesis. In a qualitative research study, the objective is to answer the research question. Because this text focuses on you as a research consumer, the following sections will primarily pertain to the evaluation of research questions and hypotheses in published research reports.

Critiquing the research question and hypothesis

The following Critical Appraisal Criteria box provides several criteria for evaluating the initial phase of the research process—the research question or hypothesis. Because the research question or hypothesis guides the study, it is usually introduced at the beginning of the research report to indicate the focus and direction of the study. You can then evaluate whether the rest of the study logically flows from its foundation—the research question or hypothesis. The author will often begin by identifying the background and significance of the issue that led to crystallizing development of the research question or hypothesis. The clinical and scientific background and/or significance will be summarized, and the purpose, aim, or objective of the study is then identified. Often the research question or hypothesis will be proposed before or after the literature review. Sometimes you will find that the research question or hypothesis is not specifically stated. In some cases, it is only hinted at or is embedded in the purpose statement, and you are challenged to identify the research question or hypothesis. In other cases, the research question is embedded in the findings toward the end of the article. To some extent, this depends on the style of the journal. Although a hypothesis can legitimately be nondirectional, it is preferable, and more common, for the researcher to indicate the direction of the relationship between the variables in the hypothesis. Quantifiable words such as “greater than,” “less than,” “decrease,” “increase,” and “positively,” “negatively,” or “related” convey the idea of objectivity and testability. You should immediately be suspicious of hypotheses or research questions that are not stated objectively. You will find that when there is a lack of data available for the literature review (i.e., the researcher has chosen to study a relatively undefined area of interest), a nondirectional hypothesis or research question may be appropriate. You should recognize that how the proposed relationship of the hypothesis or research question is phrased suggests the type of research design that will be appropriate for the study, as well as the level of evidence to be derived from the findings. Example: ➤ If a hypothesis proposes that treatment X1 will have a greater effect on Y than treatment X2, an experimental (Level II evidence) or quasi-experimental design (Level III evidence) is suggested (see Chapter 9). If a research question asks if there will be a positive relationship between variables X and Y, a nonexperimental design (Level IV evidence) is suggested (see Chapter10). Hypotheses and research questions are never proven beyond the shadow of a doubt. Researchers who claim that their data have “proven” the validity of their hypothesis or research question should be regarded with grave reservation. You should realize that, at best, findings that support a hypothesis or research question are considered tentative. If repeated replication of a study yields the same results, more confidence can be placed in the conclusions advanced by the researchers. When critically appraising clinical questions, think about the fact that the clinical question should be focused and specify the patient population or clinical problem being addressed, the intervention, and the outcome for a particular patient population. There should be evidence that the clinical question guided the literature search and that appropriate types of research studies are retrieved in terms of the study design and level of evidence needed to answer the clinical question.

Chapter 5

Introduction to qualitative research

Mark Toles, Julie Barroso

Let’s say that you are reading an article that reports findings that HIV-infected men are more adherent to their antiretroviral regimens than HIV-infected women. You wonder, “Why is that? Why would women be less adherent in taking their medications? Certainly, it is not solely due to the fact that they are women.” Or say you are working in a postpartum unit and have just discharged a new mother who has debilitating rheumatoid arthritis. You wonder, “What is the process by which disabled women decide to have children? How do they go about making that decision?” These, like so many other questions we have as nurses, can be best answered through research conducted using qualitative methods. Qualitative research gives us the answers to those difficult “why?” questions. Although qualitative research can be used at many different places in a program of research, you will most often find it answering questions that we have when we understand very little about some phenomenon in nursing.

What is qualitative research?

Qualitative research is a broad term that encompasses several different methodologies that share many similarities. Qualitative studies help us formulate an understanding of a phenomenon. Nurse scholars who are trained in qualitative methods use these methods to best answer discovery-oriented research questions. Qualitative research is explanatory, descriptive, and inductive in nature. It uses words, as opposed to numbers, to explain a phenomenon. Qualitative research lets us see the world through the eyes of another—the woman who struggles to take her antiretroviral medication, or the woman who has carefully thought through what it might be like to have a baby despite a debilitating illness. Qualitative researchers assume that we can only understand these things if we consider the context in which they take place, and this is why most qualitative research takes place in naturalistic settings. Qualitative studies make the world of an individual visible to the rest of us. Qualitative research involves an “interpretative, naturalistic approach to the world; meaning that qualitative researchers study things in their natural settings, attempting to make sense of or interpret phenomena in terms of the meaning people bring to them” (Denzin & Lincoln, 2011, p. 3).

What do qualitative researchers believe?

Qualitative researchers believe that there are multiple realities that can be understood by carefully studying what people can tell us or what we can observe as we spend time with them. Example: ➤ The experience of having a baby, while it has some shared characteristics, is not the same for any two women, and it is definitely different for a disabled mother. Thus, qualitative researchers believe that reality is socially constructed and context dependent. Even the experience of reading this book is different for any two students; one may be completely engrossed by the content, while another is reading but at the same time worrying about whether or not her financial aid will be approved soon. Because qualitative researchers believe that the discovery of meaning is the basis for knowledge, their research questions, approaches, and activities are often quite different from quantitative researchers (see the Critical Thinking Decision Path). Qualitative researchers seek to understand the “lived experience” of the research participants. They might use interviews or observations to gather new data, and use new data to create narratives about research phenomena. Thus, qualitative researchers know that there is a very strong imperative to clearly describe the phenomenon under study. Ideally, the reader of a qualitative research report, if even slightly acquainted with the phenomenon, would have an “aha!” moment in reading a well-written qualitative report. So, you may now be saying, “Wow! This sounds great! Qualitative research is for me!” Many nurses feel very comfortable with this approach because we are educated with regard to how to speak with people about the health issues concerning them; we are used to listening, and listening well. But the most important consideration for any research study is whether or not the methodology fits the question. This means that qualitative researchers must select an approach for exploring phenomena that will actually answer their research questions. Thus, as you read studies and are considering them as evidence on which to base your practice, you should ask yourself, “Does the methodology fit with the research question under study?”

Does the methodology fit with the research question being asked?

As we said before, qualitative methods are often best for helping us determine the nature of a phenomenon and the meaning of experience. Sometimes authors will state that they are using qualitative methods because little is known about a phenomenon, but that alone is not a good reason for conducting a study. Little may be known about a phenomenon because it does not matter! When researchers ask people to participate in a study, to open themselves and their lives for analysis, they should be asking about things that will help make a difference in people’s lives or help provide more effective nursing care. You should be able to articulate a valid reason for conducting a study, beyond “little is known about this topic.” Considering the examples at the start of this chapter, we may want to know why HIV-infected women are less adherent to their medication regimens, so we can work to change these barriers and anticipate them when our patients are ready to start taking these pills. Similarly, we need to understand the decision-making processes women use to decide whether or not to have a child when they are disabled, so we can guide or advise the next woman who is going through this process. To summarize, a qualitative approach “fits” a research question when the researchers seek to understand the nature or experience of phenomena by attending to personal accounts of those with direct experiences related to the phenomena. Keeping in mind the purpose of qualitative research, let’s discuss the parts of a qualitative research study

Components of a qualitative research study

The components of a qualitative research study include the review of literature, study design, study setting and sample, approaches for data collection and analysis, study findings, and conclusions with implications for practice and research. As we reflect on these parts of qualitative studies, we will see how nurses use the qualitative research process to develop new knowledge for practice

​Steps in the Research Process

• Review of the literature

• Study design

• Sample

• Setting: Recruitment and data collection

• Data collection

• Data analysis

• Findings

• Conclusions

Review of the literature

When researchers are clear that a qualitative approach is the best way to answer the research question, their first step is to review the relevant literature and describe what is already known about the phenomena of interest. This may require creativity on the researcher’s part, because there may not be any published research on the phenomenon in question. Usually there are studies on similar subjects, or with the same patient population, or on a closely related concept. Example: ➤ Researchers may want to study how women who have a disabling illness make decisions about becoming pregnant. While there may be no other studies in this particular area, there may be some on decision making in pregnancy when a woman does not have a disabling illness. These studies would be important in the review of the literature because they identify concepts and relationships that can be used to guide the research process. Example: ➤ Findings from the review can show us the precise need for new research, what participants should be in the study sample, and what kinds of questions should be used to collect the data. Let’s consider an example. Say a group of researchers wanted to examine HIV-infected women’s adherence to antiretroviral therapy. If there was no research on this exact topic, the researcher might examine studies on adherence to therapy in other illnesses, such as diabetes or hypertension. They might include studies that examine gender differences in medication adherence. Or they might examine the literature on adherence in a stigmatizing illness, or look at appointment adherence for women, to see what facilitates or acts as a barrier to attending health care appointments. The major point is that even though there may be no literature on the phenomenon of interest, the review of the literature will identify existing related studies that are useful for exploring the new questions. At the conclusion of an effective review, you should be able to easily identify the strengths and weaknesses in prior research and a clear understanding of the new research questions, as well as the significance of studying them.

Study design

The study design is a description of how the qualitative researcher plans to go about answering the research questions. In qualitative research, there may simply be a descriptive or naturalistic design in which the researchers adhere to the general tenets of qualitative research but do not commit to a particular methodology. There are many different qualitative methods used to answer the research questions. Some of these methods will be discussed in the next chapter. What is important, as you read from this point forward, is that the study design must be congruent with the philosophical beliefs that qualitative researchers hold. You would not expect to see a qualitative researcher use methods common to quantitative studies, such as a random sample, a battery of questionnaires administered in a hospital outpatient clinic, or a multiple regression analysis. Rather, you would expect to see a design that includes participant interviews or observation, strategies for inductive analysis, and plans for using data to develop narrative summaries with rich description of the details from participants’ experiences. You may also read about a pilot study in the description of a study design; this is work the researchers did before undertaking the main study to make sure that the logistics of the proposed study were reasonable. For example, pilot data may describe whether the investigators were able to recruit participants and whether the research design led them to the information they needed.

Sample

The study sample refers to the group of people that the researcher will interview or observe in the process of collecting data to answer the research questions. In most qualitative studies, the researchers are looking for a purposeful or purposively selected sample (see Chapter 10). This means that they are searching for a particular kind of person who can illuminate the phenomenon they want to study. Example: ➤ The researchers may want to interview women with multiple sclerosis or rheumatoid arthritis. There may be other parameters—called inclusion and exclusion criteria—that the researchers impose as well, such as requiring that participants be older than 18 years, not under the influence of illicit drugs, or experiencing a first pregnancy (as opposed to subsequent pregnancies). When researchers are clear about these criteria, they are able to identify and recruit participants with the experiences needed to shed light on the phenomenon in question. Often the researchers make decisions such as determining who might be a “long-term survivor” of a certain illness. In this case, they must clearly describe why and how they decided who would fit into this category. Is a long-term survivor someone who has had an illness for 5 years or 10 years? What is the median survival time for people with this diagnosis? Thus, as a reader of nursing research, you are looking for evidence of sound scientific reasoning behind the sampling plan.

When the researchers have identified the type of person to include in the research sample, the next step is to develop a strategy for recruiting participants, which means locating and engaging them in the research. Recruitment materials are usually very specific. Example: ➤ If the researchers want to talk to HIV-infected women about adherence to their medication regimen, they may distribute flyers or advertise their interest in recruiting women who consistently take their medication as indicated, as well as those who do not. Or, they may want to talk to women who fit into only one of those categories. Similarly, the researchers who are examining decision making in pregnancy among women with disabling conditions would develop recruitment strategies that identify subjects with the conditions or characteristics they want to study.

In a research report, the researcher may include a description of the study sample in the findings. (This can also be reported in the description of the sample.) In any event, besides a demographic description of the study participants, a qualitative researcher should also report on key axes of difference in the sample. Example: ➤ In a sample of HIV-infected women, there should be information about the stage of illness, what kind/how many pills they must take, how many children they have, and so on. This information helps you place the findings into a context

Setting: Recruitment and data collection

The study setting refers to the places where participants are recruited and the data are collected. Settings for recruitment are usually a point of contact for people of common social, medical, or other individual traits. In the example of HIV-infected women who are having difficulties adhering to their antiretroviral regimens, researchers might distribute flyers describing the study at AIDS service organizations, support groups for HIV-infected women, clinics, online support groups, and other places people with HIV may seek services. The settings for data collection are another critical area of difference between quantitative and qualitative studies. Data collection in a qualitative study is usually done in a naturalistic setting, such as someone’s home, not in a clinic interview room or researcher’s office. This is important in qualitative research because the researcher’s observations can inform the data collection. To be in someone else’s home is a great advantage, as it helps the researcher to understand what that participant values. An entire wall in a participant’s living room might contain many pictures of a loved one, so anyone who enters the home would immediately understand the centrality of that person in the participant’s life. In the home of someone who is ill, many household objects may be clustered around a favorite chair: perhaps an oxygen tank, a glass of water, medications, a telephone, tissues, and so on. A good qualitative researcher will use clues like these in the study setting to complete the complex, rich drawing that is being rendered in the study.

Data collection

The procedures for data collection differ significantly in qualitative and quantitative studies. Where quantitative researchers focus on statistics and numbers, qualitative researchers are usually concerned with words: what people can tell them and the narratives about meaning or experience. Qualitative researchers interview participants; they may interview an individual or a group of people in what is called a focus group. They may observe individuals as they go about daily tasks, such as sorting medications into a pill minder or caring for a child. But in all cases, the data collected are expressed in words. Most qualitative researchers use voice recorders so that they can be sure that they have captured what the participant says. This reduces the need to write things down and frees researchers to listen fully. Interview recordings are usually transcribed verbatim and then listened to for accuracy. In a research report, investigators describe their procedures for collecting the data, such as obtaining informed consent, the steps from initial contact to the end of the study visit, and how long each interview or focus group lasted or how much time the researcher spent “in the field” collecting data. A very important consideration in qualitative data collection is the researcher’s decision that they have a sufficient sample and that data collection is complete. Researchers generally continue to recruit participants until they have reached redundancy or data saturation, which means that nothing new is emerging from the interviews. There usually is not a predetermined number of participants to be selected as there is in quantitative studies; rather, the researcher keeps recruiting until she or he has all of the data needed. One important exception to this is if the researcher is very interested in getting different types of people in the study. Example: ➤ In the study of HIV-infected women and medication adherence, the researchers may want some women who were very adherent in the beginning but then became less so over time, or they may want women who were not adherent in the beginning but then became adherent; alternately, they may want to interview women with children and women without children to determine the influence of having children on adherence. Whatever the specific questions may be, sample sizes tend to be fairly small (fewer than 30 participants) because of the enormous amounts of written text that will need to be analyzed by the researcher. Investigators use great care to design the interview questions because they must be crafted to help study participants describe their personal experiences and perceptions. Interview questions are different from research questions. Research questions are typically broad, encompassing, and written in scientific language. The interview questions may also be broad, like the overview or grand tour question that seeks the “big picture.” Example: ➤ Researchers might ask, “Tell me about taking your medications—the things that make it easier, and the things that make it harder,” or “Tell me what you were thinking about when you decided to get pregnant.” Along with overview questions, there are usually a series of prompts (additional questions) that were derived from the literature. These are areas that the researcher believes are important to cover (and that the participant will likely cover), but the prompts are there to remind the researcher in case the material is not mentioned. Example: ➤ With regard to medication adherence, the researcher may have read in other studies that motherhood can influence adherence in two very different ways: children can become a reason to live, which would facilitate taking antiretroviral medication; and children can be all-demanding, leaving the mother with little to no time to take care of herself. Thus, a neutrally worded question about the influence of children would be a prompt if the participants do not mention it spontaneously. In a research report, you should expect to find the primary interview questions identified verbatim; without them, it is impossible to know how the data were collected and how the researcher shaped what was discovered in the interviews.

 EVIDENCE-BASED PRACTICE TIP Qualitative researchers use more flexible procedures than quantitative researchers. While collecting data for a project, they consider all of the experiences that may occur.

Data analysis:

Next is the description of data analysis. Here, researchers tell you how they handled the raw data, which, in a qualitative study, are usually transcripts of recorded interviews. The goal of qualitative analysis is to find commonalities and differences in the interviews, and then to group these into broader, more abstract, overarching categories of meaning, sometimes called themes, that capture much of the data. In the example we have been using about decision making regarding pregnancy for disabled women, one woman might talk about discussing the need for assistance with her friends if she became pregnant, and finding out that they were willing and able to help her with the baby. Another woman might talk about how she discussed the decision with her parents and siblings, and found them to be a ready source of aid. And yet a third woman may say that she talked about this with her church study group, and they told her that they could arrange to bring meals and help with housework during the pregnancy and afterward. On a more abstract level, these women are all talking about social support. So an effective analysis would be one that identifies this pattern in social support and, perhaps, goes further by also describing how social support influences some other concept in the data. Example: ➤ Consider women’s decision making about having a baby. In an ideal situation, written reports about the data will give you an example like the one you just read, but the page limitations of most journals limit the level of detail that researchers can present. Many qualitative researchers use computer-assisted qualitative data analysis programs to find patterns in the interviews and field notes, which, in many studies, can seem overwhelming due to the sheer quantity of data to be dealt with. With a computer-assisted data analysis program, researchers from multiple sites can simultaneously code and analyze data from hundreds of files without using a single piece of paper. The software is a tool for managing and remembering steps in analysis; however, it does not replace the thoughtful work of the researcher who must apply the program to guide the analysis of the data. In research reports, you should see a description of the way data were managed and analyzed, and whether the researchers used software or other paper-based approaches, such as using index cards with handwritten notes.

Findings: At last, we come to the results. Findings in qualitative reports, as we have suggested already, are words—the findings are patterns of any kind in the data, such as the ways that participants talked, the things that they talked about, even their behaviors associated with where the researcher spent time with them. When researchers describe patterns in the data, they may describe a process (such as the way decision making occurs); they may identify a list of things that are functioning in some way (such as a list of barriers and facilitators to taking medications for HIV-infected women); they may specify a set of conditions that must be present for something to occur (such as what parents state they need to care for a ventilator-dependent child at home); or they may describe what it is like to go through some health-related transition (such as what it is like to become the caregiver for a parent with dementia). This is by no means an all-inclusive list; rather, it is a range of examples to help you recognize what types of findings might be possible. It may help to think of the findings as discoveries. The qualitative researcher has explored a phenomenon, and the findings are a report on what he or she “found” —that is, what was discovered in the interviews and observations. When researchers describe their results, they usually break the data down into units of meaning that help the data cohere and tell a story. Effective research reports will describe the logic that was used for breaking down the units of data. Example: ➤ Are the themes—a means of describing a large quantity of data in a condensed format—identified from the most prevalent to the least prevalent? Are the researchers describing a process in temporal (time ordered) terms? Are they starting with things that were most important to the subject, then moving to less important items? As a report on the findings unfolds, the researcher should proceed with a thorough description of the phenomenon, defining each of the themes and fleshing out each of the themes with a thorough explanation of the role that it plays in the question under study. The researcher should also provide quotations that support their themes. Ideally, they will stage the quote, giving you some information about the subject from whom it came. For example, was the subject a newly diagnosed HIV-infected African American woman without children? Or was it a disabled woman who has chosen to become pregnant, but who has suffered two miscarriages? The staging of quotes is important because it allows you to put the information into some social context. In a well-written report of qualitative research, some of the quotes will give you an “aha!” feeling. You will have a sense that the researcher has done an excellent job of getting to the core of the problem. Quotes are as critical to qualitative reports as numbers are to a quantitative study; you would not have a great deal of confidence in a quantitative or qualitative report in which the author asks you to believe the conclusion without also giving concrete, verifiable findings to back it up.

Discussion of the results and implications for evidence-based practice

When the researchers are satisfied that their findings answer the research questions, they should summarize the results for you and should compare their findings to the existing literature. Researchers usually explain how these findings are similar to or different from the existing literature. This is one of the great contributions of qualitative research—using findings to open up new venues of discovery that were not anticipated when the study was designed. Example: ➤ The researchers can use findings to develop new concepts or new conceptual models to explain broader phenomena. The conceptual work also identifies implications for how findings can be used in practice and can direct future research. Another alternative is for researchers to use their findings to extend or refine existing theoretical models. For example, a researcher may learn something new about stigma that has not been described in the literature, and in writing about these findings, the researcher may refer to an existing stigma theory, pointing out how his or her work extends that theory. Nursing is a practice discipline, and the goal of nursing research is to use research findings to improve patient care. Qualitative methods are the best way to start to answer clinical and research questions that have not been addressed or when a new perspective is needed in practice. The qualitative answers to these questions provide important evidence that offers the first systematic insights into phenomena previously not well understood and often lead to new perspectives in nursing practice and improved patient care outcomes. Kearney (2001) developed a typology of levels and applications of qualitative research evidence that helps us see how new evidence can be applied to practice (Table 5.1). She described five categories of qualitative findings that are distinguished from one another in their levels of complexity and discovery: those restricted by a priori frameworks, descriptive categories, shared pathway or meaning, depiction of experiential variation, and dense explanatory description. She argued that the greater the complexity and discovery within qualitative findings, the stronger the potential for clinical application.

Findings developed with only a priori frameworks provide little or no evidence for changing practice, because the researchers have prematurely limited what they are able to learn from participants or describe in their analysis. Findings that identify descriptive categories portray a higher level of discovery when a phenomenon is vividly portrayed from a new perspective. For nursing practice, these findings serve as maps of previously uncharted territory in human experience. Findings in Kearney’s third category, shared pathway or meaning, are more complex. In this type of finding, there is an integration of concepts or themes that results in a synthesis of a shared process or experience that leads to a logical, complex portrayal of the phenomenon. The researcher’s ideas at this level reveal how discrete bits of data come together in a meaningful whole. For nursing practice, this allows us to reflect on the bigger picture and what it means for the human experience (Kearney, 2001). Findings that depict experiential variation describe the essence of an experience and how this experience varies, depending on the individual or context. For nursing practice, this type of finding helps us see a variety of viewpoints, realizations of a human experience, and the contextual sources of that variety. In nursing practice, these findings explain how different variables can produce different consequences in different people or settings. Finally, findings that are presented as a dense explanatory description are at the highest level of complexity and discovery. They provide a rich, situated understanding of a multifaceted and varied human phenomenon in a unique situation. These types of findings portray the full depth and range of complex influences that propel people to make decisions. Physical and social contexts are fully accounted for. There is a densely woven structure of findings in these studies that provide a rich fund of clinically and theoretically useful information for nursing practice. The layers of detail work together in the findings to increase understanding of human choices and responses in particular contexts (Kearney, 2001).

 EVIDENCE-BASED PRACTICE TIP Qualitative research findings can be used in many ways, including improving ways clinicians communicate with patients and with each other. So how can we further use qualitative evidence in nursing? The evidence provided by qualitative studies is used conceptually by the nurse: qualitative studies let nurses gain access to the experiences of patients and help nurses expand their ability to understand their patients, which should lead to more helpful approaches to care (Table 5.2). TABLE 5.2 Kearney’s

Modes of Clinical Application for Qualitative Research

Mode of Clinical Application

Insight or empathy: Better understanding our patients and offering more sensitive support

Assessment of status or progress: Descriptions of trajectories of illness

Anticipatory guidance: Sharing of qualitative findings with the patient

Coaching: Advising patients of steps they can take to reduce distress or improve adjustment to an illness, according to the evidence in the study

Example:

Nurse is better able to understand the behaviors of a woman recovering from depression

Nurse is able to describe trajectory of recovery from depression and can assess how the patient is moving through this trajectory

Nurse is able to explain the phases of recovery from depression to the patient and to reassure her that she is not alone, that others have made it through a similar experience

Nurse describes the six stages of recovery from depression to the patient, and in ongoing contact, points out how the patient is moving through the stages, coaching her to recognize signs that she is improving and moving through the stages.

Kearney (2001) proposed four modes of clinical application: insight or empathy, assessment of status or progress, anticipatory guidance, and coaching. The simplest mode, according to Kearney, is to use the information to better understand the experiences of our patients, which in turn helps us to offer more sensitive support.

Qualitative findings can also help us assess the patient’s status or progress through descriptions of trajectories of illness or by offering a different perspective on a health condition. They allow us to consider a range of possible responses from patients. We can then determine the fit of a category to a particular client, or try to locate them on an illness trajectory. Anticipatory guidance includes sharing of qualitative findings directly with patients. The patient can learn about others with a similar condition and can learn what to anticipate. This allows them to better garner resources for what might lie ahead or look for markers of improvement. Anticipatory guidance can also be tremendously comforting in that the sharing of research results can help patients realize they are not alone, that there are others who have been through a similar experience with an illness. Finally, coaching is a way of using qualitative findings; in this instance, nurses can advise patients of steps they can take to reduce distress, improve symptoms, or monitor trajectories of illness (Kearney, 2001). Unfortunately, qualitative research studies do not fare well in the typical systematic reviews upon which evidence-based practice recommendations are based. Randomized clinical trials and other types of intervention studies traditionally have been the major focus of evidence-based practice. Typically, the selection of studies to be included in systematic reviews is guided by levels of evidence models that focus on the effectiveness of interventions according to their strength and consistency of their predictive power. Given that the levels of evidence models are hierarchical in nature and they perpetuate intervention studies as the “gold standard” of research design, the value of qualitative studies and the evidence offered by their results have remained unclear. Qualitative studies historically have been ranked lower in a hierarchy of evidence, as a “weaker” form of research design. Remember, however, that qualitative research is not designed to test hypotheses or make predictions about causal effects. As we use qualitative methods, these findings become more and more valuable as they help us discover unmet patient needs, entire groups of patients that have been neglected, and new processes for delivering care to a population. Though qualitative research uses different methodologies and has different goals, it is important to explore how and when to use the evidence provided by findings of qualitative studies in practice.

Appraisal for evidence-based practice foundation of qualitative research

A final example illustrates the differences in the methods discussed in this chapter and provides you with the beginning skills of how to critique qualitative research. The information in this chapter, coupled with information presented in Chapter 7, provides the underpinnings of critical appraisal of qualitative research (see the Critical Appraisal Criteria box, Chapter 7). Consider the question of nursing students learning how to conduct research. The empirical analytical approach (quantitative research) might be used in an experiment to see if one teaching method led to better learning outcomes than another. The students’ knowledge might be tested with a pretest, the teaching conducted, and then a posttest of knowledge obtained. Scores on these tests would be analyzed statistically to see if the different methods produced a difference in the results.

In contrast, a qualitative researcher may be interested in the process of learning research. The researcher might attend the class to see what occurs and then interview students to ask them to describe how their learning changed over time. They might be asked to describe the experience of becoming researchers or becoming more knowledgeable about research. The goal would be to describe the stages or process of this learning. Alternately, a qualitative researcher might consider the class as a culture and could join to observe and interview students. Questions would be directed at the students’ values, behaviors, and beliefs in learning research. The goal would be to understand and describe the group members’ shared meanings. Either of these examples are ways of viewing a question with a qualitative perspective. The specific qualitative methodologies are described in Chapter 6. Many other research methods exist. Although it is important to be aware of the qualitative research method used, it is most important that the method chosen is the one that will provide the best approach to answering the question being asked. One research method does not rank higher than another; rather, a variety of methods based on different paradigms are essential for the development of a well informed and comprehensive approach to evidence-based nursing practice.

Key points •

All research is based on philosophical beliefs, a worldview, or a paradigm.

• Qualitative research encompasses different methodologies.

• Qualitative researchers believe that reality is socially constructed and is context dependent.

• Values should be acknowledged and examined as influences on the conduct of research.

• Qualitative research follows a process, but the components of the process vary.

• Qualitative research contributes to evidence-based practice.

Chapter 7

Appraising qualitative research

Dona Rinaldi Carpenter

Qualitative and quantitative research methods vary in terms of purpose, approach, analysis, and conclusions. Therefore, the use of each requires an understanding of the traditions on which the methods are based. This chapter aims to provide a set of criteria that can be used to critique qualitative research studies through a process of critical analysis and evaluation. The critical appraisal of qualitative research continues to be discussed in nursing and related health care professions, providing a framework that includes key concepts for evaluation (Beck, 2009; Bigby, 2015; Flannery, 2016; Horsburgh, 2003; Ingham-Broomfield, 2015; Pearson et al., 2015; Russell & Gregory, 2003; Sandelowski, 2015; Williams, 2015). Critical appraisal and qualitative research considerations Qualitative research represents a basic level of inquiry that seeks to discover and understand concepts, phenomena, or cultures. In a qualitative study, you should not expect to find hypotheses; theoretical frameworks; dependent and independent variables; large, random samples; complex statistical procedures; scaled instruments; or definitive conclusions about how to use the findings. A primary reason for conducting a qualitative study is to develop a theory or to discover knowledge about a phenomenon. Sample size is expected to be small. This type of research is not generalizable, nor should it be. Findings are presented in a narrative format with raw data used to illustrate identified themes. Thick, rich data are essential in order to document the rigor of the research, which is called trustworthiness in a qualitative research study. Ensuring trustworthiness in qualitative inquiry is critical, as qualitative researchers seek to have their work recognized in an evidence-driven world (Beck, 2009; Bigby, 2015). Application of qualitative research findings The purpose of qualitative research is to describe, understand, or explain phenomenon important to nursing. Phenomena are those things that are perceived by our senses. For example, pain and losing a loved one are considered phenomena. In a qualitative study, the researcher gathers narrative data that uses the participants’ voices and experiences to describe the phenomenon under investigation. Barbour and Barbour (2003) offer that qualitative research can provide the opportunity to give voice to those who have been disenfranchised and have no history. Therefore, the application of qualitative findings will necessarily be context-bound (Russell & Gregory, 2003). Qualitative research also has the ability to contribute to evidenced-based practice literature (Anthony & Jack, 2009; Cesario et al., 2002; Donnelly & Wiechula, 2013; Walsh & Downe, 2005). Describing the lived human experience of patients can contribute to the improvement of care, adding a dimension of understanding to our work as it is described by those who live it on a day-to-day basis. Fundamentally, principles for evaluating qualitative research are the same. Reviewers are concerned with the plausibility and trustworthiness of the researcher’s account of the findings and its potential and/or actual relevance to current or future theory and practice (Horsburgh, 2003; Ingham-Bloomfield, 2015; Pearson et al., 2015; Sandelowski, 2015; Williams, 2015). As a framework for understanding how the appraisal of qualitative research can support evidence-based practice, a published research report and critical appraisal criteria follow (Table 7.1). The critical appraisal criteria will be used to demonstrate the process of appraising a qualitative research report. For information on specific guidelines for appraisal of phenomenology, ethnography, grounded theory, and action research, see Chapters 5 and 6 and Streubert and Carpenter (2011). TABLE 7.1 Critical Appraisal of Qualitative Research Elements of style 1. Was there sufficient detail to enable critical appraisal? 2. Is there evidence that the researcher has the qualifications, knowledge, and expertise to conduct the research? 3. Does the abstract give a clear summary of the study, including the research problem, sample, methodology, findings, and recommendations? 4. Is the title clear, accurate, and reflective of the topic and method? Statement of the phenomenon of interest 1. Was the title clear, accurate, and related to the research question? 2. What is the phenomenon of interest and is it clearly stated for the reader? 3. What is the justification for using a qualitative method? 4. What are the philosophical underpinnings of the research method? Purpose 1. What is the purpose of the study? 2. What is the projected significance of the work to nursing? Ethical considerations 1. Is protection of human participants addressed? 2. Did the author address IRB approval? 3. Were the participants fully informed about the nature of the research? 4. Did the researcher address participant autonomy and confidentiality? Method 1. Is the method used to collect data compatible with the purpose of the research? 2. Is the method adequate to address the phenomenon of interest? 3. If a particular approach is used to guide the inquiry, does the researcher complete the study according to the processes described? Sampling 1. What type of sampling is used? Is it appropriate, given the particular method? 2. Are the informants who were chosen appropriate to inform the research? 3. Were the participants and setting adequately described and appropriate for informing the research? 4. Was saturation achieved? Data collection 1. Is data collection focused on human experience? 2. Does the researcher describe data collection strategies (i.e., interview, observation, field notes)? 3. Were the data gathered of sufficient depth and richness? 4. Were the questions asked and observations made and recorded in an appropriate way? 5. Is saturation of the data described? 6. What are the procedures for collecting data? Data analysis 1. What strategies are used to analyze the data? 2. Has the researcher remained true to the data? 3. Is there a logical connection between raw data and themes? 4. Does the reader follow the steps described for data analysis? Authenticity and trustworthiness of data 1. Does the researcher address the credibility, auditability, and transferability of the data? Credibility • Were the study purpose and method clearly described? • Do the participants recognize the experience as their own? • Has adequate time been allowed to fully understand the phenomenon? Auditability • Can the reader follow the researcher’s thinking? • Does the researcher document the research process? • Is there a logical connection between data and themes? • Is there a clear description of findings? • Is there agreement between the findings of the study and the conclusions? Transferability • Are the findings applicable outside of the study situation? • Was the selection of participants described? • Did participants fit the context of the study? • Are the results meaningful to individuals not involved in the research? • Is the strategy used for analysis compatible with the purpose of the study? Findings 1. Are the findings presented within a context? 2. Is the reader able to apprehend the essence of the experience from the report of the findings? 3. Are the researcher’s conceptualizations true to the data? 4. Does the researcher place the report in the context of what is already known about the phenomenon? Was the existing literature on the topic related to the findings? Conclusions, implications, and recommendations 1. Do the conclusions, implications, and recommendations give the reader a context in which to use the findings? 2. How do the conclusions reflect the study findings? 3. What are the recommendations for future study? Do they reflect the findings? 4. How has the researcher made explicit the significance of the study to nursing theory, research, or practice? CRITICAL APPRAISAL CRITERIA Qualitative Research Study As evidenced by published works, phenomenology is one approach to qualitative research. From a nursing perspective, qualitative research allows caregivers to understand the life experience of the patients they care for. Excerpts from “A Woman’s Experience: Living With an Implantable Cardioverter Defibrillator” by Jaclyn Conelius are provided throughout this chapter as examples of phenomenological research. The article was published in Applied Nursing Research in 2015. The following sections critique Conelius’s study. The primary purpose of this critique is to carefully examine how each step of the research process has been articulated in the study and to examine how the research has contributed to nursing knowledge. The article by Conelius (2015) provides an example of a phenomenological study true to qualitative methods. Critique of a qualitative research study The research study The study “A Woman’s Experience: Living With an Implantable Cardioverter Defibrillator” by Jaclyn Conelius, published in Applied Nursing Research, is critiqued. The article is presented in its entirety and followed by the critique. A woman’s experience: Living with an implantable cardioverter defibrillator Jaclyn Conelius, PhD, FNP-BC Abstract The implantable cardioverter defibrillators (ICD) have decreased mortality rates from those who are at risk for sudden cardiac death or who have survived sudden cardiac death and has been shown to be superior to antiarrhythmic medications (Greenburg et al., 2004). This advance in technology may improve physical health but can impose some challenges to patients, such as depression, anxiety, fear, and unpredictability. Published research on how ICD affects a woman’s life experience using phenomenology is limited. Therefore, the purpose of this article is to describe the experiences of women who have an ICD using Colaizzi’s method of phenomenology since their implant. Analysis of the three interviews resulted in five themes that described the essence of this experience. The results of this study could not only help clinicians understand what their patients are experiencing but also it can be used as an education tool. © 2014 Elsevier Inc. All rights reserved. Introduction Implantable cardioverter defibrillators (ICDs) have decreased mortality rates from those who are at risk for sudden cardiac death or who have survived sudden cardiac death and has been shown to be superior to anti-arrhythmic medications (Greenburg et al., 2004). ICDs have been supported by many clinical trials and it is now the treatment of choice in primary and secondary prevention for these patients (Bardy et al., 2005; Bristow et al., 2004; Moss et al., 2002). This mainstay of treatment has increased steadily from 486,025 implants from 2006 to 2009 to 850,068 from 2010 to 2011 (Hammill et al., 2010; Kremers et al., 2013). Of these implants approximately 28% were female only. This advance in technology may improve physical health but can impose some challenges to patients. They include the adjustments to the device in their everyday living, such as; quality of life issues as well as psychological issues. Through quantitative research the following have been reported; a fear of physical activity and a fear of shock from the device to prevent the sudden cardiac arrest (Lampert et al., 2002; Wallace et al., 2002; Whang et al., 2005). Other studies have reported anxiety, fear, and depression in these patients. Some specific fears included; malfunctioning, unpredictability, and the inability to control events (Dickerson, 2005; Dunbar, 2005; Eckert & Jones, 2002; Kamphuis et al., 2004; Lemon, Edelman, & Kirkness, 2004). These quality of life and psychological issues reported in the studies are not reported as gender specific; therefore, female specific challenges are not well studied. Furthermore, there have been few qualitative studies based on a patient’s experience of living with an ICD. Previous studies reported themes such as the feeling of gratitude, safety, belief in the future, adjustment to the device, lifesaving yet changing, fear of receiving a shock, physical/mental deterioration, confrontation with mortality and conditional acceptance (Dickerson, 2002; Fridlund et al., 2000; Kamphuis et al., 2004; Morken, Severinsson, & Karlsen, 2009; Tagney, James, & Alberran, 2003). Based on the available research studies, there is very little reported data specific to females and specifically how an ICD affects a woman’s lived experience. A lived experience is how a person immediately experiences the world (Husserl, 1970). In order to understand a woman’s lived experience living with an ICD, phenomenology was used. Phenomenology is a philosophy and a research method used to understand everyday lived experiences. Therefore, the purpose of this study was to describe what those experiences were, specifically, to describe their thoughts, feelings, and perceptions that they have experienced since their implant. It is important to gain an understanding and formulate a description of what life is for a woman who had received an implantable cardioverter defibrillator in order to describe the universal essence of that experience. Descriptive phenomenology emphasizes describing universal essences, viewing the person as one representative of the world in which she lives, an assumption of self-reflection, a belief that the consciousness is what people share and a belief that stripping of previous knowledge (bracketing) helps prevent investigator bias and interpretation bias (Wojnar & Swanson, 2007). Specifically, Colaizzi’s (1978) descriptive phenomenological method uses seven steps as a method of analyzing data so that by the end of the study a description of the lived experience could be reported. Method Descriptive phenomenology originated from the philosopher Husserl (1970), who believed that the meaning of a lived experience may be discovered though one to one interaction between the researcher and the subject. It assumes that for any human experience, there are distinct structures that make up the phenomenon. Studying the individual experiences highlights these essential structures. It is an inductive method that describes a phenomenon as it is experienced by an individual rather than by transforming it into an operationally defined behavior. An important aspect of descriptive phenomenology, according to Husserl, is the process of bracketing in which he describes as separating the phenomenon from the world and having the researcher suspend all preconceptions (Wojnar & Swanson, 2007). The goal of descriptive phenomenology is to provide a universal description of the lived experience as described by the participants of the phenomenon. Colaizzi’s (1978) method of descriptive phenomenology is the method used for this study. In his method, interviewing is the selected strategy for collecting data, which is necessary for describing an experience. This method works well with a small sample size. Sample. Ten women were asked to participate, of these, three women agreed to participate from a private cardiology office in the United States. This convenient sample of women were all Caucasian and their ages ranged from 34 to 50 years old. All three women had college degrees and have had the device over one year. None of the women were previously diagnosed with any psychiatric disease. Procedure. After receiving approval from the university’s institutional review board (IRB), women were recruited from a private cardiology office in the United States for 4 months. The participant population only included women that had an implantable cardioverter defibrillator (ICD). Women needed to be 18 years or older, and speak English. Women of all ethnic backgrounds were eligible to participate. There was no cost to the participant and no compensation provided. Once the informed consent was signed, they were asked to stay for an interview that day. All women were interviewed privately in the office and each interview lasted approximately 45 minutes to an hour. They were asked to “describe their experiences after having received an ICD, specifically, to describe their thoughts, feelings, and perceptions that they had experienced since their implant?” They were then asked to share as much of those experiences to the point that they did not have anything else to contribute. The interviews were recorded and then transcribed. The researcher conducted all of the interviews since the researcher in trained in the method. Interviews were conducted until an accurate description of the phenomenon had occurred, repetition of data and no new themes where described. This saturation of data did occurs after the three interviews. After each interview, follow up questions were asked in order to clarify any points the participant described. The researcher kept a journal to write down any notes needed during the interview. In order for the description to be pure, the researcher’s prior knowledge was bracketed to capture the essence of the description without bias (Wojnar & Swanson, 2007). Husserl (1970) introduced the term, and it means to set aside one’s own assumption and preunderstanding. In order to be true to the method, the researcher reflected and kept a journal of all assumptions, clinical experiences, understandings and biases to reference during the entire study. Significant statements and phrases pertaining to a woman’s experience living with an ICD were extracted from each transcript. These statements were written on separate sheets and coded. Meanings were formulated from the significant statements. Accordingly, each underlying meaning was coded into a specific category as it reflected an exhaustive description. Then the significant statements with the formulated meanings where grouped into themes. To ensure confidentiality, the signed informed consent forms were kept separate from the transcripts. The recorded tapes and hard copy were in a locked cabinet. Identifying information was deleted and names were never used in any research reports. Audiotapes were destroyed once the pilot study was completed. Data analysis. Each transcript was analyzed using Colaizzi’s (1978) method. The method of data analysis consisted of the following steps; (1) read all the participants’ descriptions of the phenomenon, (2) extract significant statements that pertain directly to the phenomenon, (3) formulate meanings for each significant statement, (4) categorizing into clusters of themes and validation with the original transcript, (5) describing, (6) validate the description by returning to the participant to ask them how it compares with their experience, and (7) incorporate any changes offered by the participant into the final description of the essence of the phenomenon. Rigor. There were efforts made to limit any potential bias of the researcher. One such effort was to bracket any of the researcher’s prior perspective and knowledge of the subject (Aher, 1999). To ensure the credibility of the data collected, two of the women in the study reviewed the description of the lived experiences as suggested by Lincoln and Guba (1985). This was performed as a validity check of the data. In order to address for auditability, a tape recorder was used and the researcher reviewed the transcripts and cross-referenced the field noted (Beck, 1993). Additionally, the transcripts were transcribed verbatim by a secretary in order to ensure they were free of bias. Also, the data analysis and description of the lived experience were reviewed by an independent judge with phenomenological experience to ensure intersubjective agreement. All of the themes reported were agreed upon by the judge. Finally, the researcher validated the description by returning to the participants to ask them how it compared with their experience and incorporated any changes offered by the participants into the final description of the essence of the phenomenon. This final description was reviewed by other women with ICDs who were not a part of the study to ensure fittingness. Results At the conclusion of verifying and reviewing the transcripts, there were 46 significant statements extracted that pertained directly to the phenomenon. From each significant statement formulated meanings were created. These statements were then formed into five themes (Table 1) that described the essence of these experiences. TABLE 1 Selected Examples of Significant Statements and their Formulated Meaning for Five Themes Theme NumberSignificant StatementFormulated Meaning1Security blanket: lf it keeps me alive It’s worth it.“I do not have anything to worry about anymore. I used to worry that if something happened, how soon I can get to a hospital or what could they do to try to save me.”The women did not have to worry anymore about medical emergencies.2A piece of cake: I do more than before.“Actually, I probably do a little more than before. But I can do everything that I did before. I have not eased up on anything.”She felt as if nothing has changed. She does everything she did prior.3A constant reminder: I know it’s there.“The children sometimes bump into that side and I am literally guarding that side all the time.”She is aware of it and guards it when others come in contact with it.4Living on the edge: I do not want it to go off.“I do have a little fear of that but so far, it hasn’t happened.”She has an extreme fear of the device shocking her.5Catch 22: I’d rather not have it.“I would rather not personally have it but I know medically, I need to have it, which is a good thing.”She would rather not have to have it, but she knows she needs it. Theme 1: Security blanket: If it keeps me alive it’s worth it. Women who had an ICD felt a sense of security with the device. They felt that this device acted as a security blanket. Prior to their device they had a constant worry about how soon they could get medical treatment and now that they had the device, that worry was lifted. The feeling of worry was no longer apparent for them. One woman said: Now I just think this will keep me alive long enough for somebody to make a decision, at least it will give me a chance. I do not have anything to worry about anymore. I used to worry that if something happened, how soon I could get to a hospital or what could they do to try to save me. The women also described how their worry decreased should they require medical treatment while they were with their family also was decreased. “Now I do not have to worry if I am with my family, I have ICD in my chest to give me treatment right away.” Another woman felt that the device just being there saved her life. “If the device can save her life it’s worth it.” The device prevents the heart from having sustained lethal arrhythmias. She explained: “I feel like it saved my life, I feel like it keeps my heart beating nice and smooth.” There was an overall feeling that the device improved their lives. Based on their past medical history, the device was needed since it is the next step in their medical treatment. All the women were glad they were able to receive the device. One woman explained: It could be both ways. I mean, I feel knowing what my family history is, yeah, I am glad I have it. I needed it. It made me feel that I can go anywhere and do anything because it acts like my insurance policy. Theme 2: A piece of cake: I do more than before. The women did not have a decrease in physical functioning or quality of life. Their quality of life remained stable or improved once the post operative period was over. One woman explained: Actually, I probably do a little more than before. But I can do everything that I did before. I have not eased up on anything. I felt like after the surgery, I was tired for 2 days then I could go on and do everything I used to do; now I do not even think about it. I just go about my day as usual and even do more because I know I have this to protect me. The women felt that the whole process of receiving an ICD was easy. Nothing much changed in their everyday lives. They live and do everything that they did before with no restrictions. Another woman shared, After that, I really have had no change in lifestyle. My life has been as normal as it was before. Physically, I see no change, or even see an improvement. Theme 3: A constant reminder. I know it’s there. The women felt as if they had a constant reminder of the ICD. Their family was aware of the device in their body since they can see the scar. Some family members would comment on the device if they could feel it when given a hug. This in turn would remind the women that it was there. The device did affect their body image; it made them more conscious of the device in their chest. One woman with school aged children explained: And it is hard when the kids cuddle up to me and I have to say I can’t have you on my left side anymore. With four kids, you know the pile up, at least the two youngest ones, they want to lie next to me while watching TV or when we are praying or reading books or doing anything. I have to remind them that you can’t put your head up there. The children sometimes bump into that side and I am literally guarding that side all the time. The most amount of pain that women had experienced was postoperative. After that, it varied when the pain decreased. The actual incision is “hardly noticeable” in all of the women although the knowledge that the device is in fact in their chest is a “constant reminder.” The degree at which it reminds them varies depending on body type. One woman stated: “I am reminded of this all the time, I can feel it, I know it is there. Everyday activities like opening a jar, it pops and moves. Anytime I use my pectoral muscle, I know it is there, which is a lot of what I do during the day, like laundry.” Another woman stated: “The only thing that bothers me a little bit sometimes, it feels like it moves in my chest when I am in bed. When I lay a certain way it sometimes feels like it is popping out or something.” Yeah, I mean just being that it is there and it should not be there and it shows itself all the time. I especially know it’s there in the summer when you were fewer clothes, especially bathing suits. To me it is constant reminder that I may feel fine, but I am technically sick. Theme 4: Living on the edge. I do not want it to go off. All of the women had a common fear that was constantly in their thoughts. They feared that the device would have to do its job; it would have to “fire.” They did not want this fear to become a reality. They feared that they would be somewhere in public and the device would have to administer therapy or shock them. The women stated such things as: I do have a little fear of that but so far, it hasn’t happened. Oh! I don’t want it to go off! I am completely scared it will go off and no one will know what the heck happened. The fear of the device firing has a significant impact on these women. The most concerning part, is the wonder on what it will actually feel like, the uncertainty. These women could not possibly know how it would feel like since none of them have ever received a shock. They have been told that it feels like an “animal kicking you in the chest.” None of them to date have yet to experience it. To them that is unimaginable until it becomes a reality. I am scared. I am afraid it is going to kick off and I was told it would feel like a pair of boots kicking you in the chest. And I am afraid, but it has never gone off. You know, I am wondering what it would feel like. The doctor explained it almost like getting kicked in the chest by a horse. Well, that would be a jolt, I guess? I am afraid that I will be doing something, not feel anything, then all of a sudden boom! Theme 5: Catch 22: I’d rather not have it. The women received these ICDs because it was medically necessary for them to have it based on the current guidelines. They have various cardiac medical conditions that require an implant of a defibrillator. The women understood that it was essential and yet they would rather not have had to go through it. They would rather not have the heart disease that comes with needing the device. I would rather not personally have it but I know it is medically, I need to have it, which is a good thing that I have it. Mentally it bothers me, mentally; I know I cannot avoid it. The women felt that the experience was depressing. They were mostly depressed immediately preceding the implantation. Although, it had decreased over time, there was a constant reminder of the device still there. They needed to adjust to the device, which was hard for them. They felt as if they had no choice to adjust to this new situation. One woman explained: Well, I have adjusted to it, I had no choice. But in another aspect, no, I would rather not be going through this. Interestingly, no one has ever asked me how I feel about having one before. I just got it and the doctor does not even ask me about it. I mean it comes and goes, because a lot of things I know are happening are like, it could get depressing. I do feel anxious at times, then I feel depressed at times, then I am fine at times. So, I guess it depends on what is going on. Discussion Aspects of the five themes that describe the essence of a woman’s experience living with an ICD have been reported in previous studies, but nowhere is there a study that is an exact comparison to this study. For instance, theme 1 (security blanket: if it keeps me alive it’s worth it) is similar to the concept in Fridlund et al. (2000), a feeling of gratitude, and a feeling of safety. The women in this study expressed a feeling of safety and appreciation since they received their ICDs. This sense of safety and trust in the device is consistent with other studies (Bilge et al., 2006; Dickerson, 2002; Morken et al., 2009). Contrary to what is found in the literature, the women in this study reported how they have more energy than before and noticed an actual increase in physical functioning. Previous studies have identified decreased physical functioning (Dickerson, 2005; Kamphuis et al., 2004; Williams, Young, Nikoletti, & McRae, 2007) and a decrease in activity levels in their day-to-day lives (Bolse, Hamilton, Flanangan, Caroll, & Fridlund, 2005; Eckert & Jones, 2002). This contradiction can be related to the types of studies conducted. Previous studies have used questionnaires while this study focused on actual descriptions experienced by participants who had undergone the device implant. Theme 3 (a constant reminder: I know it’s there) described the women “knowing that the device was in their chest,” and it was a reminder of their condition. They also described how it affected their body image. There were two other studies that had mentioned this as a concern for women. One study by Walker et al. (2004) reported body image concerns of women. The women in that study were more concerned on how the device appeared in their chest (i.e. the scar) than any other aspect. A second study by Tagney et al. (2003), also reported body image concerns in women since it can be seen in their chest which makes them aware of the device. There were similarities with respect to body image only. They were not concerned with the constant reminder aspect of the cardiac disease, only a constant reminder of their mortality (Dickerson, 2002). The common concern as described in theme 4 (Living on the Edge: I do not want my device going off) was the fear of the device having to shock them as well as the uncertainty of when, where, and who would be around for support. This was foremost in their thoughts. There have been common themes of fear of the device going off or shocking them in the literature reviewed. Dickerson (2002, 2005) reported that uncertainty of when and where shocks can be triggered was a prevailing concern of the male and female participants. Also, participants in Albarran, Tagney, and James (2004) study reported a feeling of uncertainty regarding the device firing. The prevailing concern in theme 5 (catch 22: I’d rather not have it.) is the conflict women have after receiving a device. These women knew that they medically needed the device yet would have rather not have gone through with it. Dickerson (2005) reported the theme of conditional acceptance that touches on the same concept. Also, a greater acceptance of the new situation was reported in previous studies (Carroll & Hamilton, 2005; Kamphuis et al., 2004). The women in this study offered specific experiences of living with an ICD which is not completely seen in any previous study as stated previously. Moreover, there were some similar aspects identified in other studies such as receiving a shock and feeling of safety but most were not specific to women (Bilge et al., 2006; Dickerson, 2002, 2005; Morken et al., 2009). This study was able to describe the essence of women who are living with an ICD. As stated previously, the majority of the patients who receive ICD s are male and all of the samples in previous studies have been predominantly male. This study is specific to women and allows special insight to women who are living with cardiac disease and more specifically cardiac disease requiring a medical device. Clinical implications and future research This study can have an impact on clinical practice as a whole by helping clinicians understand what their patients are experiencing. The women in this study stated that they experienced a lot of uncertainty regarding the need for the device and its functionality. This uncertainty can be reduced or eliminated by educating the patients with respect to how the device operates. An increase in education pre and post operatively on device functionality would benefit patients by relieving some of that uncertainty. These concerns are not being addressed properly in the healthcare system. This study can help clinicians gain the understanding of the experience these women are having and perhaps pay closer attention to these issues when they are seen in outpatient settings. Furthermore, this study can also advocate for support groups for women. Support groups would allow these patients to converse with other women with the same health condition. There are multiple studies in the literature regarding the use of support groups in heart failure patients, however, there are very few studies involving patients with ICDs. Support groups can expose women to different types of resources in order to cope better, decrease anxiety and answer any questions that arise (Myers & James, 2008). Also, it would give them a security knowing that they would be able to have each other as a support system. The women in this study were similar in that they were Caucasian from affluent areas with numerous resources available to them (Smeulders et al., 2010). An additional study involving women of various ethnical backgrounds and ages would allow capture of a wider range of experiences. Also, since the women have an outstanding fear of the device firing/shocking them, a noteworthy follow-up study would be to describe their experience post firing/shock. These studies would help clinicians understand what their patients are experiencing. lt would allow them to be more empathetic and identify the gaps in knowledge. The results would become a valuable teaching tool to help educate patients regarding their device function. The critique This is a critical appraisal of the article, “A Woman’s Experience: Living With an Implantable Cardioverter Defibrillator” (Conelius, 2015) to determine its usefulness and applicability for nursing practice. Abstract The purpose of the abstract is to provide a clear overview of the study and summarize the main features of the findings and recommendations. The abstract should accurately represent the remainder of the article. Conelius (2015) summarized the research in the following narrative: The implantable cardioverter defibrillators (ICD) have decreased mortality rates from those who are at risk for sudden cardiac death or who have survived sudden cardiac death and has been shown to be superior to antiarrhythmic medications (Greenburg et al., 2004). This advance in technology may improve physical health but can impose some challenges to patients, such as depression, anxiety, fear, and unpredictability. Published research on how an ICD affects a woman’s life experience using phenomenology is limited. Therefore, the purpose of this article is to describe the experiences of women who have an ICD using Colaizzi’s method of phenomenology since their implant. Analysis of the three interviews resulted in five themes that described the essence of this experience. The results of this study could not only help clinicians understand what their patients are experiencing but also it can be used as an education tool. Introduction/review of literature All research requires the investigator to review the literature. This is the point at which gaps are identified with regard to what is known about a particular topic and what is not known. In qualitative research, the literature review is generally brief, because there is not a great deal known about the topic; nor is there an existing body of research studies. This essentially means that the researcher needs to have an understanding of the substantive body of knowledge on the topic and a clear perspective of what areas still need to be explored. A clear rationale for why the research is needed should be established. The researcher must be clear that a gap in nursing knowledge was identified, there is a clear need for the study, and the selected research method is appropriate. Bracketing what is known about the phenomenon is one way to prevent bias and keep what is known about the topic separate, prior to data collection and analysis (see Chapter 6). Conelius (2015) discusses bracketing in the data collection section of her research on women and implantable cardiac defibrillators. The background information provided in her introduction establishes a need for a qualitative study. Conelius (2015) emphasizes the fact that to date much of the research has been quantitative. She further notes that qualitative studies to date have not been gender specific, emphasizing the need for a study related to women’s experiences. Implantable cardioverter defibrillators (ICDs) have decreased mortality rates from those who are at risk for sudden cardiac death or who have survived sudden cardiac death and has been shown to be superior to anti-arrhythmic medications (Greenburg et al., 2004). ICDs have been supported by many clinical trials and it is now the treatment of choice in primary and secondary prevention for these patients (Bardy et al., 2005; Bristow et al., 2004; Moss et al., 2002). This mainstay of treatment has increased steadily from 486,025 implants from 2006 to 2009 to 850,068 from 2010 to 2011 (Hammill et al., 2010; Kremers et al., 2013). Of these implants approximately 28% were female only. (Conelius, 2015) This advance in technology may improve physical health but can impose some challenges to patients. They include the adjustments to the device in their everyday living, such as; quality of life issues as well as psychological issues. Through quantitative research the following have been reported; a fear of physical activity and a fear of shock from the device to prevent the sudden cardiac arrest (Lampert et al., 2002; Wallace et al., 2002; Whang et al., 2005). Other studies have reported anxiety, fear, and depression in these patients. Some specific fears included; malfunctioning, unpredictability, and the inability to control events (Dickerson, 2005; Dunbar, 2005; Eckert & Jones, 2002; Kamphuis et al., 2004; Lemon, Edelman, & Kirkness, 2004). These quality of life and psychological issues reported in the studies are not reported as gender specific; therefore, female specific challenges are not well studied. Furthermore, there have been few qualitative studies based on a patient’s experience of living with an ICD. Previous studies reported themes such as the feeling of gratitude, safety, belief in the future, adjustment to the device, lifesaving yet changing, fear of receiving a shock, physical/mental deterioration, confrontation with mortality and conditional acceptance (Dickerson, 2002; Fridlund et al., 2000; Kamphuis et al., 2004; Morken, Severinsson, & Karlsen, 2009; Tagney, James, & Alberran, 2003). Based on the available research studies, there is very little reported data specific to females and specifically how an ICD affects a woman’s lived experience. (Conelius, 2015) Phenomenology is a philosophy and a research method used to understand everyday lived experiences and is an appropriate methodology for the phenomena of interest. The subjective experience of women with an ICD is central to study and key to developing interventions to help these women cope. Conelius (2015) clearly articulates the focus of the study and makes a clear case for why a qualitative design is appropriate. When critiquing the literature review of a qualitative study, it is important to remember that this component of the study must be critiqued within the context of the qualitative methodology selected. In phenomenological studies, the literature review may be delayed until the data analysis is complete in order to minimize bias. Conelius (2015) does not indicate that the review was delayed. Philosophical underpinnings In addition to making a case for the study and qualitative approach, it is also important to give the reader perspective on the philosophical traditions of the method selected. Conelius (2015) describes the philosophical underpinnings of phenomenology and then relates the traditions to the method used in the study. In most published studies, the author is most concerned about sharing the findings of the study. This limits the space for in-depth literature reviews or discussion of the method used. Conelius (2015) discusses the work of Husserl (1970) as being an integral component of her philosophical grounding of phenomenology as method. She then connects this fundamental work to the method developed by Colaizzi (1978). A lived experience is how a person immediately experiences the world (Husserl, 1970). In order to understand a woman’s lived experience living with an ICD, phenomenology was used. Phenomenology is a philosophy and a research method used to understand everyday lived experiences. Descriptive phenomenology emphasizes describing universal essences, viewing the person as one representative of the world in which she lives, an assumption of self-reflection, a belief that the consciousness is what people share and a belief that stripping of previous knowledge (bracketing) helps prevent investigator bias and interpretation bias (Wojnar & Swanson, 2007). Specifically, Colaizzi’s (1978) descriptive phenomenological method uses seven steps as a method of analyzing data so that by the end of the study a description of the lived experience could be reported. (Conelius, 2015) The specific qualitative research approach selected helps determine the focus of the research and the manner in which sampling, data collection, and analysis are undertaken. The qualitative research example provided here used phenomenology as method. Research studies using a qualitative approach other than phenomenology should be critiqued relative to the philosophical underpinnings of the method. Purpose The author explained why the study was important and the significant contribution the study would make to nursing’s body of knowledge. The background information justified the use of a qualitative approach as well as why phenomenology was used. The researcher states that “The purpose of this study was to describe a woman’s experience living with an ICD. More specifically to describe their thoughts, feelings and perceptions that they have experienced since their implant” (Conelius, 2015). The purpose is clearly articulated, first in the abstract and then in the introduction of the study. Conelius (2015) makes it clear that there is a gap in nursing knowledge related to ICDs and the experience of women living with an ICD. Ethical considerations Addressing the ethical aspect of a research report involves being able to know whether participants were told what the research entailed, how their autonomy and confidentiality were protected, and what arrangements were made to avoid harm. In qualitative research the data collection tools generally include interview and participant observation, making anonymity impossible. Because the interviews are open-ended, the possibility of disclosing personal information or uncomfortable experiences related to the topic may occur. Consent must be a process of continuous negotiation (Oye et al., 2016). The study by Conelius (2015) was approved by the Institutional Review Board. The author clearly states how the participants were protected. “To ensure confidentiality the signed informed consent forms were kept separate from the transcripts. The recorded tapes and hard copy were in a locked cabinet. Identifying information was deleted and names were never used in any research reports. Audiotapes were destroyed once the pilot study was completed” (Conelius, 2015). Participants were fully informed about the nature of the research and were protected from harm; their autonomy and confidentiality were protected. Conelius (2015) also made clear to the participants that they had the right to withdraw from the research at any time. This is true for any research; however, in a qualitative investigation, ethical issues may arise at any point in the study (Hegney & Chan, 2010). Conelius (2015) clearly articulated the ethical rigor of this study. Sample In qualitative research, participants are recruited because of their life experience with the phenomena of interest. This is referred to as purposeful sampling. The goal is to ensure rich, thick data about the phenomenon of interest. Data are generally collected until no new material is emerging and data saturation has been reached. Cleary and colleagues (2014) discuss sampling in qualitative research in relationship to sample size. Qualitative studies generally have a small sample. Following the steps for sampling in qualitative research, Conelius (2015) offers the following information related to participant selection: After receiving approval from the university’s institutional review board (IRB), women were recruited from a private cardiology office in the United States for 4 months. The participant population only included women that had an implantable cardioverter defibrillator (ICD). Women needed to be 18 years or older, and speak English. Women of all ethnic backgrounds were eligible to participate. There was no cost to the participant and no compensation provided. Once the informed consent was signed, they were asked to stay for an interview that day. (Conelius, 2015) In qualitative research, purposive sampling is the approach of choice. Participants must have experience with the phenomenon of interest and be appropriate to inform the research. In this case, Conelius (2015) needed women with an ICD. Her selection process supports a qualitative sampling paradigm that is appropriate for phenomenology. Data generation The data generation approach should be sufficiently described so that it is clear to the reader why a particular strategy was selected. Conelius (2015) clearly articulates that the data generation method supports a qualitative paradigm and allows for discovery, description, and understanding of the participants’ lived experience. The researcher uses open-ended questioning and asks each individual to exhaust their ideas and describe their experiences. She also completes three in-depth interviews with each participant, allowing for clarification of responses as well as an opportunity for the participants to add experiences that may have been omitted at the first interview. Recording and transcribing the interview verbatim helps maintain authenticity of the data. The following excerpts from the article illustrate these points: All women were interviewed privately in the office and each interview lasted approximately 45 minutes to an hour. They were asked to “describe their experiences after having received an ICD, specifically, to describe their thoughts, feelings, and perceptions that they had experienced since their implant?” They were then asked to share as much of those experiences to the point that they did not have anything else to contribute. The interviews were recorded and then transcribed. The researcher conducted the interviews since the researcher was trained in the method. Interviews were conducted until an accurate description of the phenomenon had occurred, repetition of data and no new themes where described. This saturation of data did occur after the three interviews. (Conelius, 2015) The researcher kept a journal to write down any notes needed during the interview. “In order for the description to be pure, the researcher’s prior knowledge was bracketed to capture the essence of the description without bias (Wojnar & Swanson, 2007). Husserl (1970) introduced the term, and it means to set aside one’s own assumption and preunderstanding. In order to be true to the method, the researcher reflected and kept a journal of all assumptions, clinical experiences, understandings and biases to reference during the entire study. Significant statements and phrases pertaining to a woman’s experience living with an ICD were extracted from each transcript. These statements were written on separate sheets and coded. Meanings were formulated from the significant statements. Accordingly, each underlying meaning was coded into a specific category as it reflected an exhaustive description. Then the significant statements with the formulated meanings where grouped into themes.” (Conelius, 2015) Data generation was appropriate for this study and followed the steps described by Colaizzi (1978). Data analysis The process of data analysis is fundamental to determining the credibility of qualitative research findings. Data analysis involves the transformation of raw data into a final description or narrative, identifying common thematic elements found in the raw data. The description should enable a reviewer to confirm the processes of concurrent data collection and analysis as well as steps in coding and identifying themes. Data analysis followed the method described by Colaizzi (1978). The author developed a table to allow the reader to follow the line of thinking and establish thematic elements. The reader can clearly follow the researcher’s stated processes. Further, Conelius (2015) followed clear processes to establish authenticity and trustworthiness of the data. The findings reported demonstrate the participants’ realities. During data analysis the researcher made every effort to eliminate potential bias. Bracketing, verbatim transcription of taped interviews, and an independent reviewer were used to establish intersubjective agreement. Authenticity and trustworthiness Critical to the meaning of the findings is the researcher’s ability to demonstrate that the data were authentic and trustworthy or valid. Rigor ensures there is a correlation between the steps of the research process and the actual study. Procedural rigor relates to accuracy of data collection and analysis. Rigor or trustworthiness is a means of demonstrating the credibility and integrity of the qualitative research process (Cope, 2014). A study’s rigor may be established if the reviewer is able to audit the actions and development of the researcher. It is at this point that the review of literature becomes critical and should be systematically related to the findings. This was addressed by the author, and every effort was clearly employed to reduce any bias or misinterpretation of findings. Conelius (2015) was able to demonstrate rigor with regard to data analysis in multiple ways. She stated: There were efforts made to limit any potential bias of the researcher. One such effort was to bracket any of the researcher’s prior perspective and knowledge of the subject (Aher, 1999). To ensure the credibility of the data collected, two of the women in the study reviewed the description of the lived experiences as suggested by Lincoln and Guba (1985). This was performed as a validity check of the data. In order to address for auditability, a tape recorder was used and the research was reviewed the transcripts and cross-referenced the field noted (Beck, 1993). Additionally, the transcripts were transcribed verbatim by a secretary in order to ensure they were free of bias. The data analysis and description of the lived experience were reviewed by an independent judge with phenomenological experience to ensure intersubjective agreement. All of the themes reported were agreed upon by the judge. Finally, the researcher validated the description by returning to the participants to ask them how it compared with their experience and incorporated any changes offered by the participants into the final description of the essence of the phenomenon were created. Conelius (2015) provided clear evidence of rigor for the reader. Bracketing, having participants read the final description and thematic elements, taping and transcribing interviews verbatim, and using an independent judge to establish intersubjective agreement are key elements in a well done qualitative study. The author also left an audit trail illustrated in table format. This table establishes the researcher’s line of thinking. Examples of how raw data lead to the identification of thematic elements were provided and further establish rigor for this study. Findings, conclusions, implications, and recommendations Findings from a qualitative study generally are discussed in a narrative format that tells the story of the experience through an exhaustive description and thematic elements. Conelius (2015) summarized conclusions, implications, and recommendations from the study. The findings were also compared to prior research studies. In qualitative research, this is the area that must include a comprehensive incorporation of current research on the topic. According to Conelius (2015): Aspects of the five themes that describe the essence of a woman’s experience living with an ICD have been reported in previous studies, but nowhere is there a study that is an exact comparison to this study. For instance, theme 1 (security blanket: if it keeps me alive it’s worth It) is similar to the concept in Fridlund et al. (2000), a feeling of gratitude, and a feeling of safety. The women in this study expressed a feeling of safety and appreciation since they received their ICDs. This sense of safety and trust in the device is consistent with other studies. (Bilge et al., 2006; Dickerson, 2002; Morken et al., 2009) Contrary to what is found in the literature, the women in this study reported how they have more energy than before and noticed an actual increase in physical functioning. Previous studies have identified decreased physical functioning (Dickerson, 2005; Kamphuis et al., 2004; Williams, Young, Nikoletti, & McRae, 2007) and a decrease in activity levels in their day-to-day lives (Bolse, Hamilton, Flanangan, Caroll, & Fridlund, 2005; Eckert & Jones, 2002). This contradiction can be related to the types of studies conducted. Previous studies have used questionnaires while this study focused on actual descriptions experienced by participants who had undergone the device implant. Theme 3 (a constant reminder: I know it’s there) described the women “knowing that the device was in their chest,” and it was a reminder of their condition. They also described how it affected their body image. There were two other studies that had mentioned this as a concern for women. One study by Walker et al. (2004) reported body image concerns of women. The women in that study were more concerned on how the device appeared in their chest (i.e., the scar) than any other aspect. A second study by Tangney et al. (2003), also reported body image concerns in women since it can be seen in their chest which makes them aware of the device. There were similarities with respect to body image only. They were not concerned with the constant reminder aspect of the cardiac disease, only a constant reminder of their mortality. (Dickerson, 2002) The common concern as described in theme 4 (Living on the Edge: I do not want my device going off) was the fear of the device having to shock them as well as the uncertainty of when, where, and who would be around for support. This was foremost in their thoughts. There have been common themes of fear of the device going off or shocking them in the literature reviewed. Dickerson (2002, 2005) reported that uncertainty of when and where shocks can be triggered was a prevailing concern of the male and female participants. Also, participants in Albarran, Tagney, and James’ (2004) study reported a feeling of uncertainty regarding the device firing. The prevailing concern in theme 5 (catch 22: I’d rather not have it.) Is the conflict women have after receiving a device. These women knew that they medically needed the device yet would have rather not have gone through with it. Dickerson (2005) reported the theme of conditional acceptance that touches on the same concept. Also, a greater acceptance of the new situation was reported in previous studies. (Carroll & Hamilton, 2005; Kamphuis et al., 2004) The women in this study offered specific experiences of living with an ICD which is not completely seen in any previous study. Moreover, there were some similar aspects identified in other studies such as receiving a shock and feeling of safety but most were not specific to women. (Bilge et al., 2006; Dickerson, 2002, 2005; Morken et al., 2009) This study was able to describe the essence of women who are living with an ICD. The study remained true to qualitative research design. The focus on women was important, as there have been no gender specific studies to date. Capturing the fear and uncertainty for women with an ICD can have an impact on clinical practice and patient education. The author emphasized that these concerns are not being addressed properly in the healthcare system. This study can help clinicians gain an understanding of the experience these women are having and perhaps pay closer attention to these issues when they are seen in outpatient settings (Conelius, 2015). The research may also be helpful in the establishment of support groups for women with ICDs. “Support groups can expose women to different types of resources in order to cope better, decrease anxiety, and answer any questions that arise” (Myers & James, 2008). “Since the women have an outstanding fear of the device firing/shocking them, a noteworthy follow-up study would be to describe their experience post firing/shock” (Conelius, 2015). By capturing the experiences of women with ICDs, the potential for better sensitivity toward the patient experience exists. This may be critical to overall quality of life and extends beyond the actual purpose and operation of the device. Conelius (2015) has made an important contribution to the understanding of women’s experiences with an ICD. The critical appraisal of a qualitative study involves an in-depth review of each step of the research process. The example of a qualitative critique in this chapter provides a foundation for the development of critiquing skills in qualitative research. Go to Evolve at http://evolve.elsevier.com/LoBiondo/ for review questions, critiquing exercises, and additional research articles for practice in reviewing and critiquing.