_

Research in Nursing & Health, 2001, 24, 145±153

Focus on Research Methods

Levels and Applications of Qualitative Research Evidence

Margaret H. Kearney�

Boston College School of Nursing, 140 Commonwealth Avenue, Chestnut Hill, MA 02467-3812

Received 8 May 2000; accepted 4 December 2000

Abstract: Evaluation of qualitative ®ndings for application to nursing prac- tice can go beyond the rigor with which the evidence was developed to the characteristics of the ®ndings themselves. Five categories of qualitative ®ndings are described that vary 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 explana- tory description. Four modes of clinical application of qualitative evidence are proposedÐinsight or empathy, assessment of status or progress, anti- cipatory guidance, and coachingÐthat vary in their degree of visibility and patient involvement. The greater the complexity and discovery within quali- tative ®ndings, the stronger may be the potential for clinical application. ß 2001 John Wiley & Sons, Inc. Res Nurs Health 24: 145±153, 2001

Keywords: qualitative research; evidence-based practice

Most health care practice is based on some kind of evidence, but rarely is it based exclusively on outcomes of randomized controlled trials. In fact, most judgments by practitioners draw on a complex array of formal and informal informa- tion sources, shuf¯ed and sorted based on the salient characteristics of the situation at hand. Published research ®ndings are one formal infor- mation source, and qualitative research is an important subset. Yet little has been written about how to transfer the speci®c kinds of information produced in qualitative research to the interactive arena of health care delivery. The goal of this article is to offer for discussion two preliminary frameworks: one for classifying and the other for using qualitative research ®ndings.

The Role of Qualitative Research Evidence in Nursing Practice

Although the use of nursing research ®ndings to direct practice decisions is as old as Florence

Nightingale's work in the Crimea, there has been a recent surge in attention to evidence-based nurs- ing practice (Mulhall, Alexander, & le May, 1998; Simpson & Knox, 1999), in which the ideal is to select procedures for patient assessment and care delivery based on experimental evidence that one approach is better than another. Several cautionary voices have emerged in this discus- sion. Although impressive bodies of research have been built to guide nursing approaches to certain clinical problems, experimental evidence to support many common nursing care concerns is patchy at best (Mitchell, 1999). The minting of the randomized controlled trial as the gold standard of evidence discounts other in¯uential knowledge sources, such as colleagues, salient experiences, and ethical and clinical judgment (Estabrooks, 1998), that nurses and others use in their therapeutic interactions with patients.

Human caring, the central landscape of much nursing practice, is less easily quanti®ed than are surgical procedures or pharmacological regi- mens (Benner & Wrubel, 1989; Mitchell, 1999).

*Associate Professor.

ß 2001 John Wiley & Sons, Inc. 145

However, qualitative health research produces knowledge situated in the intra- and interpersonal realm: what health and illness feels like to patients; where interpretations of health and illness experience come from; how experiences change under a variety of interpersonal, his- torical, cultural, and other conditions; and how the subtleties of human in¯uence, including professional in¯uence, can propel or derail ill- ness adjustment and recovery (Green & Britten, 1998).

According to Stetler (1994), utilization of research ®ndings by individuals or groups can be instrumental (concrete applications in practice protocols), conceptual (cognitive application through new insights and understandings of situations), or symbolic (use of ®ndings to legitimate a policy or practice approach). As Sandelowski (1997), Estabrooks (1998), and others have observed, qualitative evidence for practice likewise can be used in instrumental forms in clinical assessment, protocols, and policies but is perhaps especially well suited to conceptual uses, in which by reading qualitative ®ndings nurses gain access to the experiences and observed actions of patients and others and thereby expand their stores of theoretical under- standing, which reveal more helpful approaches to care.

In order to use qualitative ®ndings as evidence for either conceptual or instrumental purposes, nurses need frameworks for evaluation and com- parison of the methods and ®ndings of qualitative studies. Much has been written on standards for adequate conduct of qualitative research (Lincoln & Guba, 1985). To defend the systematic and thorough nature of their work against quantitative expectations for reliability and validity, qualita- tive researchers have focused more on defending the rigor with which the research was conducted than on the usefulness of the ®ndings themselves. Both quantitative and qualitative ®ndings are assumed worthy if achieved systematically and without bias. Yet qualitative ®ndings can be systematic and unbiased but less than illuminat- ing for human learning or clinical practice in a particular situation. The following discussion identi®es two characteristics of qualitative ®nd- ingsÐcomplexity and discoveryÐthat can be found in varying degrees across methodological approaches. Although exceptions certainly do exist, the current study suggests, in agreement with others (Swanson, Durham, & Albright, 1997), that the higher the levels of complexity and discovery, the greater is the potential for clinical insight and application.

Degrees of Complexity and Discovery in Qualitative Findings

The utility of research ®ndings in speci®c situa- tions must be based not only on their ®t with the clinical issues at hand but also on the richness and informativeness of the ®ndings as evidence. At their best, qualitative ®ndings teach the reader something about how context, history, and indi- viduality constitute meaning and explicate human action in a closely observed, highly speci®c unique situation. Such a research report can be a gold mine for clinical insights. Less well integrated or narratively vivid presentations may offer less to clinicians in search of ideas for practice.

A way of characterizing the richness of infor- mation in qualitative ®ndings is in terms of complexity and discovery. Complexity is de®ned here as the substantiated linking of discrete ®ndings into a multifaceted web of interactions. Components of ®ndings that may be integrated in this way include aspects of the context of a phenomenon, such as historical, familial, eco- nomic, social, environmental, and political in¯u- ences, and aspects of human individuality and experience, such as perception, meaning, emo- tion, action, and interaction. Discovery is de®ned as the presentation by researchers of new perspectives on or information about the human phenomenon under study. New perspectives or information may be revealed, for example, in verbatim accounts that portray the experience under study for the ®rst time or with previously uncaptured richness, or in a theoretical or inter- pretive framing of the phenomenon that sheds light on how it came to be and what it is like. Although newness depends on the knowledge and perspective of the reader, for this purpose it connotes a characterization of the phenomenon under study not previously commonly described or not accepted as known within the public discourse in the discipline.

The qualitative methods commonly applied by nurse researchersÐcontent analysis, phenomen- ology, ethnography, and grounded theoryÐpro- duce differing degrees of complexity when applied to their full potential. In general, des- criptive methods, such as qualitative descrip- tion (Sandelowski, 2000) and certain forms of descriptive phenomenology, are designed to pro- duce a lower level of complexity, and interpretive and theorizing methods, such as hermeneutic phenomenology and grounded theory, aim for a higher level. Nonetheless, a high degree of discovery is possible with all these methods. For

146 RESEARCH IN NURSING & HEALTH

example, qualitative description is intended to portray subjective experience of a phenomenon without complex interpretations or theoretical linkages by the researcher, but it can achieve discovery in bringing to light fresh perspectives from participants. On the other hand, a visibly complex `̀ grounded'' theory may not discover new knowledge about a phenomenon if the theory in fact replicates other published work on the topic in similar samples and settings.

For this discussion ®ve categories of com- plexity and discovery in qualitative evidence are suggested, four of which have clinical utility (the ®rst category lacks discovery and hence provides no new clinical guidance). These groupings can apply both to original work and to syntheses of qualitative ®ndings. Although some may question the use of hierarchical evaluation systems as devaluing more basic and exploratory forms of inquiry, that is not the intention here. As sug- gested above, simply structured ®ndings are not inherently of lower quality than more complex ®ndings, but they are descriptively different. Either may be arrived at using strong and rigorous methods and be useful for a given clinical situation. Examples cited here were drawn from several volumes of a convenient and reputable source, Qualitative Health Research, in order to reduce the confusion of methodological rigorÐ well monitored by the peer reviewers of this journalÐwith complexity and discovery. The intent here is to elucidate the differences in structure in ®ve types of qualitative ®ndings and discuss the impact of these differences when ®ndings are considered as evidence for practice.

Findings restricted by a priori frameworks. Findings that are produced by applying an existing set of ideas to qualitative data without identifying new insights or enriching, extending, or revising existing theory may offer a certain degree of complexity but little discovery, and consequently they provide little or no evidence for practice. This restriction occurs when a researcher has claimed to have described a phenomenon using inductive techniques, but the ®ndings have been visibly constrained by a dominant predetermined framework. In effect, data are collected and then ®tted to an externally imposed model without close attention to new evidence that these data might contain.

Restriction of supposedly inductive ®ndings by inappropriately applied theoretical frameworks must be distinguished from extension and enrich- ment of existing theory. Indeed, Sandelowski (1993) has noted that although the role of theory varies across qualitative approaches, atheoretical

qualitative work is impossible and undesirable. Disciplinary and philosophical worldviews una- voidably shape the goals of research and how ®ndings are interpreted by nurses, sociologists, educators, and others. If qualitative ®ndings are set loose in the research literature without integration into the matrix of knowledge in the ®eld, little is gained for clinical application.

By contrast, discovery is aborted when a researcher sets out to analyze data with the goal of inductive discovery, sees a resemblance to a popular theory, and abandons the inductive process in favor of categorizing new data in old bottles, so to speak. Although such reports are rarely found in major research journals that have qualitative-methods experts on their review panels, in other venues in the nursing literature they are more common. For example, well-known works on women's `̀ relatedness'' and `̀ connec- tions'' (Gilligan, 1982; Miller & Stiver, 1997) have inspired many a reiteration of these concepts in manuscripts and dissertations without clear substantiation in the data, preventing the dis- covery of instances where relatedness is not central, or other insights might be offered. Like- wise, a popular concept from nursing scholarship (uncertainty, transition, coping, or the like) may be identi®ed as the major descriptor of a pheno- menon under study, but the careful reader ®nds that verbatim data provided in support of this claim reveal alternative views that have not been acknowledged in the analysis. If the primary marker of excellence in qualitative work is to `̀ privilege subjective meaning or lay knowledge,'' as Popay, Rogers, and Williams (1998, p. 344) put it, then one should proceed with caution when overlaying an existing theory on participants' efforts to express their situated and personal views. Clinicians seeking understanding of human experiences or responses will ®nd little to work with if the analyst has obscured rather than illuminated those experiences.

Descriptive categories. The simplest level of complexity of qualitative evidence, but one that nonetheless may portray discovery, is a series of labeled data categories. Clusters of data are labeled with brief headings that indicate the topic or type of data contained therein, as in a book's table of contents. Content analysis techniques often produce ®ndings in this format. Areas of participant or researcher concern are indicated by the data clusters, but linking of concepts to each other or of experience to context is not detailed by the analyst, and the ®ndings are not complex structure-wise. When diagrammed, the categories may be linked in a wheel or daisy format, indi-

QUALITATIVE EVIDENCE / KEARNEY 147

cating that all categories of data are components of a central idea, but the speci®c nature of these relationships is not detailed.

A high level of discovery may be achieved in descriptive categories when a phenomenon is portrayed from a new perspective through the presentation of vivid and informative data in a clear and helpful set of categories. In this form the data speak for themselves. For example, in a study entitled, `̀ My Hurts,'' Woodgate and Kristjanson (1996) categorized children's descri- ptions of pain, producing a comprehensive list of pain descriptors, attributed causes, and children's views on good care during painful episodes. These data, logically organized, provide a vivid immersion in the children's views. In a second example, Engebretson (1996) analyzed observa- tions and interviews of clients and healers who used healing touch therapy. Experiential descrip- tions were categorized into physical sensations, emotional experience, and visual experience, the labels of which were simple but whose content was informative. Steinberg, Davila, Collazo, Loew and Fischgrund (1997) described attitudes, perceptions, and beliefs of Hispanic families with deaf children and presented clinically useful portraits of the families' views on causality, their responses and that of their community, ways of communicating with the child, and adequacy of services.

Discovery of previously undescribed aspects of experience is demonstrated in these examples, although the structure of these ®ndings is simple. This level of complexity ®ts well with initial exploratory work when the goal is to begin to describe an experience, and it can provide stimulus for future research. Descriptive categori- zations can suggest the breadth of patients' worries or the scope of their learning needs, information that can have great clinical utility. They can serve as maps of previously uncharted landscapes in human experience.

Shared pathway or meaning. The third cate- gory shows an increase in complexity. Here, the investigator's interpretation produces a synthesis of a shared experience or process. It is distin- guished from the previous grouping by the investigator's integration of concepts or themes into a linked and logical portrayal. The common- ality, core, or essence of the experience is captured by the analyst. This synthesis has the potential to reveal something previously un- described about the phenomenon that would not be readily apparent in a series of unlinked cate- gories. The increased complexity enables greater discovery.

Shared pathways or meanings can be seen in ®ndings achieved with a variety of qualitative approaches. For example, in a qualitative descrip- tion, data clusters are linked in a holistic picture of the experience. In a grounded-theory study, concepts are connected in a model of in¯uences and strategies or actions, with each relationship substantiated by data. In a phenomenological study, themes are experiential components inte- grated into a narrative depiction of a multifaceted phenomenon. The analyst has moved from des- cribing parts of a data set to explaining how these are components of a larger social or experiential whole.

Examples here include Schreiber's (1996) description of women's process of recovery from depression. Women were shown to move through six phases, and the conditions for progression are clearly described. Each phase was fully ¯eshed out with evidence of its origins in the preceding phase. Bott, Cobb, Scheibmeir, and O'Connell (1997) described the salient themes in the ex- perience of quitting smoking, including the inten- sity of the struggle, the personi®cation of the cigarette, the planning process and how this time was different, and how it was necessary to replace the old habit with a new habit as a transi- tional measure. Marcus (1998) portrayed a four- stage process by which women moved toward addiction recovery in a therapeutic community, each stage entailing several kinds of work on the self, moving toward preparation for a new world- view and life role. Barroso (1997) developed four dimensions of reconstructing one's life as a long- term survivor of AIDS, the dimensions forming a `̀ web of meaning; if one were to pull a strand from the web, the whole structure is likely to collapse'' (p. 63).

Those perusing these research reports move beyond reading the quotes to reading the analyst's ideas. The investigator's interpretation has shown how discrete data bits come together in a mean- ingful whole, allowing re¯ection on the larger picture and what it means for human experience and health care clients.

Depiction of experiential variation. An even greater degree of qualitative complexity is achieved in ®ndings that not only describe the main pathway or essence of an experience but also portray how that experience or pathway varies depending on individuality and context. Portraying or explaining variation in a human experience requires considerable breadth and depth of sampling and data collection and a high level of analytic expertise, which can pro- duce a high degree of discovery of new insights

148 RESEARCH IN NURSING & HEALTH

or perspectives on human phenomena. Although grounded theories are the type of ®ndings most commonly thought of as portraying variation, this level of complexity can be seen in other approaches as well. The rich detail in a fully realized phenomenology or ethnography can cap- ture a variety of viewpoints and realizations of a human experience and the contextual sources of that variety, whether political, cultural, familial, or intrapersonal.

For example, in a grounded-theory study Wilson, Hutchinson, and Holzemer (1997) des- cribed how men with AIDS salvaged their quality of life in a context of cultural, sexual, and linguistic diversity, depicted in a multivariate model. For each stage of the basic psychosocial process, details were given on the variations in its expression and in the conditions necessary for progression, explaining why some moved toward satisfactory life quality, whereas others could not. These ®ndings are at a higher level of complexity than those of Barroso (1997), intro- duced above, in which family reaction to HIV was the only variation described. Likewise, in contrast to the portrayal by Marcus (1998) of ad- dicted women's shared pathway toward recovery, Pursley ± Crotteau and Stern (1996) described how pregnant women recovering from cocaine addiction might or might not move forward to create a new life depending on how much struc- ture they imposed on their behavior and how much they desired to give up drugs and change their lives. The researchers diagrammed this relationship in a four-quadrant model. Condi- tional models (in which several `̀ variables'' vary and thereby produce different consequences) are hallmarks of this level of fully realized grounded theory.

Experiential variation also can be demon- strated in narrative descriptions such as that of Harris (2000), a recounting of young women's self-harmÐof `̀ cutting the bad out of me''Ð which vividly evokes the situations under which this painful experience worsened or was relieved and how the responses of others could in¯uence its progression or remission. Studies of structures and organizations also can portray and explain variation, as exempli®ed by the analysis of King, Stewart, King, and Law (2000) of the organiza- tional characteristics and issues affecting the longevity of self-help groups for parents of special needs children. By observing six groups and speaking with a range of parents, they were able to identify the qualities of groups that succeeded versus those that were not able to sustain them- selves over time. If the answer to the research

question is `̀ It depends,'' the ®ndings have reached this level.

Dense explanatory description. The ®fth and highest level of complexity and discovery might be termed the qualitative gold standard. These ®ndings may be seen as representing the chara- cteristics of the `̀ experiential variation'' category when achieved to the highest degree, but in doing so, these ®ndings appear qualitatively different. Clifford Geertz (1973) exempli®ed this level of ®ndings in his de®nition of `̀ thick description,'' but this level of explanatory detail can be seen outside ethnography as well. These ®ndings are a rich evocation of a situated understanding of a multifaceted and varied human phenomenon in a unique situation. With dense factual and des- criptive detail, they portray the full depth and range of complex in¯uences that propel persons to make one choice over another, to speak one way versus another, and to view life one way rather than another. Physical and social context is colorfully conveyed, and the experience of time is captured at the levels of social, cultural, and political history as well as at the level of personal progress. The role of the researcher in this particular context and interaction also is clearly apparent.

Dense explanatory description may be achieved using a number of qualitative methods includ- ing ethnography, phenomenology, and grounded theory. The researcher provides excursion into the participants' evolving lives as shaped by a constellation of historical and cultural in¯uences as well as by the life experience of the researcher. In addition to feeling a feeling or gaining a deep grasp of the forces leading to a decision, the reader learns previously unrevealed unique qua- lities of a particular set of events in a particular historical and cultural milieu. These ®ndings contribute to or extend the theory, whether expli- citly or by demonstration, in portraying a dynamic by which a phenomenon can be framed across situations.

Most dense explanatory descriptions are book length, and, indeed, qualitative studies realized at this level may be portrayed less fully when constrained by journal format (Sandelowski, 1997). In journal article form they focus on a small piece of a highly particular experience and illuminate it exquisitely. For example, Davis and Joakimson (1997) examined the folk ill- ness of `̀ nerves'' in two different yet similar cultural contexts: ®shing villages in Norway and Newfoundland. Despite many parallels in culture, access to medical care, and social construction of illness, the researchers demonstrated that the

QUALITATIVE EVIDENCE / KEARNEY 149

social milieu and social consequences of having nerves differed greatly in these two communities based on differences in the longevity and liveli- hood of the ®shing industry and in the strength and cohesion of women's social networks. These conditions produced contrasting social views of nerves as status or stigma, as public or private, as originating in the psyche or the soma, and as expression of belonging or of alienation.

In a second example Tourigny (1998) captured the unique cultural conditions and familial break- down and hopelessness in a decaying inner city in the Midwest that led to `̀ some new dying trick: African-American youths `choosing' HIV/AIDS'' (p. 149). Physical violence, gang membership, loss of social structure and purpose, and the demands of caretaking in families where because of disease or drugs the parents were unable to function as adults led several young people to deliberately expose themselves to HIV as a grasp for meaning and purpose in life.

In a third study Wilson, Morse, and Penrod (1998) looked intensively at how caregiving relationships were built over the course of a single week in a summer camp for ventilator- dependent children. They used photography and thorough observations, the latter of which pro- duced ®eld notes with many action examples although only a few verbatim comments. The investigators recognized the unique context of their study, in which there was abatement of the parental contingencies resulting from illness such as worry about the future and the rest of the family, thus allowing caregiving to ¯ourish in a normalized atmosphere. The progression of physical, verbal, and affective communication between a primary professional caregiver and a child was mapped in exquisite detail. The actions that enabled a dyad to progress from a tentative relationship to a connected relationship were depicted in ®eld notes, in which the role of the observer in the scene as well as the context of the interaction was described. There was explora- tion of variations such as whether a particular camper was new or returning and whether a care- giver was new or known. The ®t of this process within a larger body of theory-building work on caregiving was discussed at length.

Although they arose from different disciplinary contexts and describe a range of human phenom- ena, the densely woven structure of these ®ndings enabled them to have in common the discovery of a rich fund of clinically and theoretically useful evidence, in which layers of detail work together to increase understanding of human choices and responses in particular contexts. Not all quali-

tative analysis achieves this level, and not all qualitative work needs to do so to meet its goals. Nonetheless, appreciation of the distinct nature of such extraordinary contributions is warranted. They provide a wealth of evidence of human experience that can be mined to enrich health care practice.

Applying Qualitative Evidence in Health Care Encounters

Qualitative ®ndings, both complex and simple, that portray newly discovered concepts or expla- nations of human experience can be applied directly in certain clinical situations without intervening steps of replication or transformation into quantitative tools. The following is an ex- ploration of ways of using qualitative evidence in clinical encounters and of the role of complexity and discovery in this application, offered in the hope of advancing dialogue between research and practice.

Qualitative evidence can be applied in at least four ways of increasing degrees of visibility and patients' involvement. Qualitative ®ndings in a range of complexity levels can serve as evidence for practice if discovery is presentÐthat is, if new information about a phenomenon has been revealed. The higher the complexity level of the qualitative evidence, the more information can be gleaned on how and when to apply it (Swanson et al., 1997). When ®ndings portray a worldview with great vividness and explain how different contexts affect a health experience, their rele- vance and ®t with a given situation are better able to be judged than if given only a list of quotes or concepts without contextual settings or insightful interpretation. The mode of application depends on the complexity level of the evidence; the particularities of the patient situation, which determine the ®t and utility of the evidence; and, most important, the clinician's judgment about what that patient needs.

Insight or empathy. The simplest mode of applying qualitative evidence is using the infor- mation to better understand the health experience, as suggested in the conceptual mode of research utilization referred to by Stetler (1994) and Sandelowski (1997). Clinicians can learn from qualitative ®ndings such things as what it feels like to be in a given illness situation, the common factors affecting individuals as they consider health care options, and the different ways of looking at a particular illness. They can use this learning to better grasp the nature of the patient's

150 RESEARCH IN NURSING & HEALTH

experience and to offer support in a more sensi- tive way. This insight can be gained from any level of qualitative ®ndings that have quotations or description vivid enough to evoke vicarious experience.

Using this understanding, the clinician pays attention to new cues from the patient, recognizes typical or atypical responses, makes sense of comments or behaviors previously considered inconsistent, and is empowered to reach out with support in a more informed way than before. Simply knowing what a health experience has felt like to a particular group of study participants can be extremely instructive. One can keep such evidence in mind, compare it to patients' des- criptions and behaviors, and enrich one's stock of knowledge over time without mentioning it to the patient if the clinician judges this would not be helpful. The usual toolbox of clinical judgment skills must be used here and in all applications in order to gauge whether the information from a study report really does make sense for an individual patient in her or his stage of illness and cultural and social context. Fittingness (Beck, 1993) in a given clinical setting does not originate in the ®ndings themselves but only within the professional's carefully considered clinical judg- ment of all the evidence at hand.

Assessment of status or progress. Many qualitative ®ndings, from descriptive categories to higher levels of complexity, suggest a trajec- tory of illness experience or describe different perspectives on a particular health condition. Formal clinical assessment tools can be deve- loped from qualitative ®ndings and tested in larger samples, but using qualitative ®ndings to consider a range of possible responses or points on a trajectory for an individual patient also is worthy of consideration. If one has read, for example, that there may be three ways of reacting to the diagnosis of diabetes or of behaving as fathers-to-be during labor, one can use clinical cues and skilled questions to determine the ®t of a category to a particular client or family, with its probable orientation and associated responses. In this way a clinician gains a set of possibilities for clinical exploration, problems to watch out for, and emotions and behaviors to understand. Likewise, if it is believed there is a clinical ®t of a client's situation with study ®ndings that include a trajectory, such as stages of recovery from cardiac events or steps to reach adjustment after a traumatic loss, it is possible to mentally locate one's patient on that progression and based on that comparison to make judgments as to possible problems with or the speed of recovery for

that patient. These observations are silent and, as always, ®ltered among many other clinical judgments, but they have the potential to speed the recognition of exemplary recovery or the knowledge of when a client is `̀ stuck.''

Anticipatory guidance. This mode includes open sharing of qualitative ®ndings with clients. As such, it can be considered more interventionist, yet perhaps more cautious while still empower- ing, in the sense the patient is brought into the discussion of ®t and relevance of qualitative ®nd- ings to her or his situation. In this approach clinicians share openly with patients what qualita- tive ®ndings suggest that the patients may ®nd themselves experiencing or the stages that may be ahead. Patients are offered a research-based perspective on what they may be going through, based on how other people (that is, study participants) have described it. This usually requires evidence at the level of shared pathway or meaning or on higher degrees of complexity. The purpose of sharing ®ndings in a clinical dialogue is to help clients anticipate such factors as what resources may be needed to deal with obstacles ahead or what markers are linked to feeling better. It also provides patients an oppor- tunity to feel less alone in an illness experience or, if the ®t is not apparent, to offer previously undisclosed but clinically relevant information about their own experience.

Qualitative research offers much to patients in its systematic depiction of human experience. Its scope surpasses the personal sagas of celebrities found in paperback bookstores or the advice of peers in self-help groups. Clinicians are obliged to describe the sources of qualitative health information and their limitations and to serve as editors and translators of methods and ®ndings, just as when discussing results of randomized trials. Nonetheless, clients deserve exposure to peer-reviewed qualitative research information just as they deserve access to health information sources on the Internet and in the popular press, although they may be more limited or misleading.

Coaching. In this most active mode of apply- ing qualitative evidence, clinicians share qualita- tive ®ndings and advise patients of steps they might take to reduce distress or improve adjust- ment based on that evidence. If qualitative evidence has shown that diabetics were best able to make peace with blood-sugar-control regimens when those prescribed regimens were adapted to their own lifestyles, when they worked coopera- tively with clinicians in a team approach, and when they paid attention to both biochemical and embodied signs of changes or imbalances, then

QUALITATIVE EVIDENCE / KEARNEY 151

based on this evidence, one might advise selected patients that doing these things might make them fare better. Coaching patients based on qualita- tive evidence is most secure when a topic has available a body of qualitative ®ndings with a moderate- to high degree of complexity. Evidence that portrays or explains variation is invaluable here because the higher the level of complexity of the qualitative evidence, the more clearly will one be able to portray the contingencies that may affect outcomes and the range of possibilities in the experience at hand.

There will be no odds ratios or relative risks to offer a client in support of the recommenda- tionÐonly the power of a diligent systematic study of human experience in a particular con- text. Although sharing qualitative ®ndings with patients in the process of anticipatory guidance and coaching may seem like new and treacherous territory in the increasingly outcomes-driven pra- ctice arena, it is little different from the continual application of the experiential storehouse of exemplars of former patients (and colleagues' stories about their former patients), which drives many clinical assessments and conclusions. Involving the patient in making a connection with reported health-related experiences enables validation of the clinician's hunches; demon- strates to clients the privileging of subjective accounts (Popay et al., 1998) like their own; and increases the likelihood of open communication, mutual goal setting, and true collaboration.

DISCUSSION

This exploration has been intended as a ®rst step toward continued dialogue on the role of qualitative ®ndings in improving nursing care. It is hoped that the concepts of complexity and discovery, the categories of ®ndings, and the modes of clinical application described here will soon be expanded, revised, or challenged by other writers. Certainly, there are qualitative ®ndings that straddle levels or have qualities of more than one level. Others who use these groupings to assess the studies cited as examples may arrive at different conclusions. In not addressing rigor or the faithful conduct of qualitative methods, this article lets stand the extant approaches to determining whether qualitative evidence is credible. In addition, no delving has been done into how critics should most constructively respond to the body of published work that claims to have reached a higher level of com- plexity than it actually has. Nor has this article

addressed whether these kinds of qualitative evidence apply to historical research, action research, discourse analysis, or a number of other qualitative approaches. Further dialogue on these topics is anticipated.

Clinical acumen rooted in much thought and experience is needed to use qualitative evidence (and any evidence) in practice. Barriers to clinical application include mistrust of qualitative meth- ods by practitioners and standard-setters (but not necessarily patients); lack of time or motivation to seek out and learn ®ndings of qualitative studies in one's area of practice; fatigue from decoding the awkward and unnatural language of some qualitative researchers; and pressure within the practice context to limit time with the patient and limit care to standardized assessments and tech- nical and pharmacological interventions.

Qualitative researchers can facilitate clinical application by conducting qualitative data collec- tion and analysis that is as deep and rich as needed to answer the research question and that makes the best possible use of participants' time and energy. To maximize discovery, data collec- tors must enable participants to tell their stories or show their lives in detail and in their own frame- works and according to their own priorities, and they use methods to their full potential: in pheno- menology, making time and space for repeated iterations of experience from different perspec- tives and engaging in deep levels of immersion; in grounded theory, doing intensive analysis between interviews or observations so that new questions are asked and the theory moves forward with each contact; in ethnography, staying in the ®eld long enough and with enough participation to be able to describe the border crossing of one's own acculturated, situated body into a new history, culture, and worldview. In data analysis and writing, researchers can maximize the utility of ®ndings by striving for the highest possible level of complexity and discovery. Portraying or explaining experiential variation and ®nding evocative language seem to be the most dif®cult steps for novices. Shaping ®ndings into narrative can be helped by studying writers of all kinds, including journalists, playwrights, and ®ction writers. Reading ®ction can be a good course of study to learn how to balance detail with sweep, action with dialogue, and description with interpretation.

In discussing their ®ndings, researchers can provide detailed and contextually speci®c sug- gestions as to how a clinician might use them (Johnson, 1997), with guidance for identifying appropriate patients and settings. Researchers

152 RESEARCH IN NURSING & HEALTH

might identify important types of experiences not represented in their studies that might be expected to differ, or they might suggest ques- tions clinicians could ask to determine the ®t of the evidence with a patient's situation. With fully drawn ®ndings and guidance from qualita- tive researchers, clinicians can add qualitative research evidence to their experiential, anecdotal, quantitative, and other knowledge sources, com- bining relative risks with portraits of situated experience when crafting individualized patient care.

REFERENCES

Barroso, J. (1997). Reconstructing my life: Becoming a long-term survivor of AIDS. Qualitative Health Research, 7, 57 ± 74.

Beck, C.T. (1993). Qualitative research: Evaluation of its credibility, ®ttingness, and auditability. Western Journal of Nursing Research, 15, 263 ± 266.

Benner, P., & Wrubel, J. (1989). The primacy of caring: Stress and coping in health and illness. Menlo Park, CA: Addison ± Wesley.

Bott, M., Cobb, A., Scheibmeir, M., & O'Connell, K. (1997). Quitting: Smokers relate their experiences. Qualitative Health Research, 7, 255 ± 269.

Davis, D., & Joakimson, L. (1997). Nerves as status and nerves as stigma: Idioms of distress and social action in Newfoundland and northern Norway. Qualitative Health Research, 7, 370 ± 390.

Engebretson, J. (1996). Urban healers: An experiential description of American healing touch groups. Qualitative Health Research, 6, 526 ± 541.

Estabrooks, C. (1998). Will evidence-based nursing practice make practice perfect? Canadian Journal of Nursing Research, 30(1),15 ± 26.

Geertz, C. (1973). The interpretation of cultures: Selected essays. New York: Basic Books.

Gilligan, C. (1982). In a different voice: Psychological theory and women's development. Cambridge, MA: Harvard University Press.

Green, J., & Britten, N. (1998). Qualitative research and evidence-based medicine. British Medical Journal, 316, 1230 ± 1232.

Harris, J. (2000). Self-harm: Cutting the bad out of me. Qualitative Health Research, 10, 164 ± 173.

Johnson, J. (1997). Generalizability in qualitative research. In J. Morse (Ed.), Completing a qualitative project: Details and dialogue (pp. 191 ± 208). Thou- sand Oaks, CA: Sage.

King, G., Stewart, D., King, S., & Law, M. (2000). Organizational characteristics and issues affecting the longevity of self-help groups for parents of children with special needs. Qualitative Health Research, 10, 225 ± 241.

Lincoln, Y., & Guba, E. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage.

Marcus, M. (1998). Changing careers: Becoming clean and sober in a therapeutic community. Qualitative Health Research, 8, 466 ± 480.

Miller, J.B., & Stiver, I. (1997). The healing conne- ction: How women form relationships in therapy and in life. Boston: Beacon Press.

Mitchell, G. (1999). Evidence-based practice: Critique and alternative view. Nursing Science Quarterly, 21, 30 ± 35.

Mulhall, A., Alexander, C., & le May, A. (1998). Appraising the evidence for practice: What do nurses need? Journal of Clinical Effectiveness, 32, 54 ± 58.

Popay, J., Rogers, A., & Williams, G. (1998). Rationale and standards for the systematic review of qualita- tive literature in health services research. Qualitative Health Research, 8, 341 ± 351.

Pursley ± Crotteau, S., & Stern, P. (1996). Creating a new life: Dimensions of temperance in perinatal crack cocaine users. Qualitative Health Research, 6, 350 ± 367.

Sandelowski, M. (1993). Theory unmasked: The uses and guises of theory in qualitative research. Research in Nursing & Health, 16, 213 ± 218.

Sandelowski, M. (1997). `̀ To be of use'': Enhancing the utility of qualitative research. Nursing Outlook, 45, 125 ± 132.

Sandelowski, M. (2000). Whatever happened to quali- tative description? Research in Nursing & Health, 23, 334 ± 340.

Schreiber, R. (1996). (Re)de®ning my self: Women's process of recovery from depression. Qualitative Health Research, 6, 469 ± 491.

Simpson, K.R., & Knox, G.E. (1999). Strategies for developing an evidence-based approach to perinatal care. MCN, 24, 122 ± 132.

Steinberg, A., Davila, J., Collazo, J., Loew, R., & Fischgrund, J. (1997). `̀ A little sign and a lot of love . . .'': Attitudes, perceptions, and beliefs of Hispanic families with deaf children. Qualitative Health Research, 7, 202 ± 222.

Stetler, C.B. (1994). Re®nement of the Stetler/Marram model for application of research ®ndings to practice. Nursing Outlook, 42, 15 ± 25.

Swanson, J., Durham, R., & Albright, J. (1997). Cli- nical utilization/application of qualitative research. In J. Morse (Ed.), Completing a qualitative project: Details and dialogue (pp. 253 ± 281). Thousand Oaks, CA: Sage.

Tourigny, S. (1998). Some new dying trick: African American youths `̀ choosing'' HIV/AIDS. Qualita- tive Health Research, 8, 149 ± 167.

Wilson, H., Hutchinson, S., & Holzemer, W. (1997). Salvaging quality of life in ethnically diverse patients with advanced HIV/AIDS. Qualitative Health Research, 7, 75 ± 97.

Wilson, S., Morse, J., & Penrod, J. (1998). Developing reciprocal trust in the caregiving relationship. Qualitative Health Research, 8, 446 ± 465.

Woodgate, R., & Kristjanson, L. (1996). `̀ My hurts'': Hospitalized young children's perceptions of acute pain. Qualitative Health Research, 6, 184 ± 201.

QUALITATIVE EVIDENCE / KEARNEY 153