The Practice of Clinical Autonomy in Hospitals: 20 000 Nurses Tell Their Story

In both the 2001 study on the dimensions of magnetism and the 3 studies to identify structures, interviewees provided hundreds of examples and descriptions in response to requests and questions such as the following: • Describe a situation in which you functioned autonomously.

Abstract

How can autonomous practice affect nurses' professional job satisfaction?In a previous article,1 we described how staff nurses, managers, and physicians working on units with confirmed healthy work environments judged "competent performance of nurses." Although related to all of the essentials of a healthy work environment,2 competent performance is a sine qua non for autonomous decision making, the essential professional work process we discuss in this article.Autonomy has long been cited as 1 of the 3 cornerstones of excellent, magnetic work environments.3 Progress in identifying organizational structures and best practices that enable clinical autonomy has been inhibited by widespread confusion, lack of precise definition of clinical autonomy, and failure to distinguish between organizational and clinical autonomy. The following excerpt4 from groups of staff nurses illustrates this confusion:In 2001, staff nurses in 14 magnet hospitals identified 8 essentials of a healthy (ie, job satisfying and professionally productive) work environment,2 1 of which is clinical autonomy. Six studies related to the 8 essentials of magnetism have been conducted (Table 1). In both the 2001 study2,4–7 on the dimensions of magnetism and the 3 studies1,10–16 to identify structures, interviewees provided hundreds of examples and descriptions in response to requests and questions such as the following:Through constant comparative analysis,18 we generated grounded theories from which we constructed the magnet hospital staff nurses' definition of autonomy and the items for the Clinical Autonomy subscale of the Essentials of Magnetism (EOM) tool. In the 2 psychometric studies8,9,17 summarized in Table 1, respondents from both magnet and comparison hospitals completed the EOM; the construct validity of the EOM was established by comparing the scores of nurses in these 2 groups of hospitals. Nurses in magnet hospitals consistently reported healthier work environments, including opportunity and support for autonomous clinical practice, than did staff nurses in comparison hospitals.8,17This article is not based on a single research study. It is a synthesis of the results from 6 studies1,2,4–17 on the essentials of magnetism conducted from 2001 to 2007 as they pertain to clinical autonomy and the results of an informal survey at the 2006 National Training Institute in Anaheim, California, of critical care nurses about their perceptions of autonomous practice. Unless noted otherwise, all excerpts are from interviews with staff nurses in the 2001 study2,4–7 or in the 2004 to 2007 structure-identification studies.1,10–16 All the speakers are staff nurses unless noted otherwise. Suggestions offered are specifically addressed to clinical nurses and what these nurses might do to improve their work environment with respect to autonomous practice.Staff nurses perceive and report that they feel that they should practice autonomously, that it is expected of them, but that they receive little support for doing so.19 In the 15 years from 1974 to 1991, nurses reported only low to moderate autonomy scores.19,20 Part of the problem may be due to inadequate, faulty, inconsistent measurement. When autonomy is measured by means of items such as "Nurses need more autonomy in their daily practice,"21 it is impossible to know which of the 34 different definitions of autonomy in the literature is being used as a referent.12 Hence, the results cannot be interpreted accurately. However, when results from 3 different studies19,22,23 by 3 different investigating teams of 3 different samples of staff nurses, some of whom were nurses in magnet hospitals, in which the same instrument22 was used to measure autonomy were compared, the level of autonomy was only moderate; there was little change in level of autonomy during 20 years and little support for autonomous practice. This moderate level and lack of support for autonomous practice were also evident in a 2001 interview study2,4 of 279 staff nurses in 14 magnet hospitals. When we asked nurses to describe a situation in which they functioned autonomously, 39% provided examples and descriptions indicating limited, unsanctioned, unsupported, or no autonomy.4 By 2004, more than 100 hospitals, 80% of them magnet hospitals, had been tested by using the EOM containing the Clinical Autonomy subscale. Although autonomy scores in magnet hospitals were significantly higher than those in comparison hospitals, the scores still were only moderately high.12 In 2003, the mean autonomy score for magnet hospitals was 78.59, 70% of the total possible score; in 2006, the mean score was 76.38, 68% of the total possible score.8,17No word engenders more misunderstanding, confusion, and differences in conceptualization than does the word autonomy. Six different descriptors—clinical, job/work, professional, individual, practice, and organizational—are used, and, to add to the confusion, autonomy is dynamic; it changes over time. As an experienced nurse said, "In the 1980s, refusing to give a patient a contraindicated drug was an act of heroism; in the 1990s, it was an example of autonomy; today, it's standard practice."Staff nurse interviewees had a clearer and more consistent understanding of what constitutes clinical autonomy than what is indicated by the descriptions and definitions in the literature.4,12 They demonstrated this by their high similarity in examples of autonomy, descriptions of steps and components of the autonomy process, and illustrations of impediments to autonomous practice. When staff nurses use the word autonomy, they mean clinical autonomy; sometimes, clinical autonomy is also termed practice or professional practice autonomy. Staff nurses do not group clinical autonomy with control of practice and ability to self-govern as was done by directors of nursing in 1982 and has continued in much of the literature today.24 Today's magnet hospitals staff nurses and their Canadian counterparts25 clearly distinguish the self-determination, self-regulation, control of practice characteristics of a profession from the autonomy characteristic, that is, "the freedom to make decisions about the service needs of clients" as defined by Flexner.26The following definition of autonomy was constructed through constant comparative and thematic analysis18 and from the grounded theory of autonomy generated from the examples and descriptions provided by staff nurses interviewed in hospitals all across the United States:Autonomous practice includes both types of decision making—independent and interdependent.Understanding the concept of unique and overlapping (U/O) spheres of practice and their relationship to type of decision making (independent or interdependent) is essential for safe, effective autonomous practice. U/O spheres may also be termed separate and combined. The knowledge and activities of nurses and the knowledge and activities of physicians or other health professionals can be envisioned as 2 or more partly intersecting spheres (see FigureF1). The more discrete and separate knowledge and activities of each professional are indicated by the parts of the sphere that do not intersect. Those that are shared are indicated by the overlapping areas.27 Usually health maintenance, prevention, and caring dominate nurses' unique sphere, whereas curative, diagnostic, and prescriptive functions dominate the physicians' unique sphere. Professionals make independent, patient-centered decisions in their unique spheres of practice. Nurses may seek counsel and advice from peers, but the decision making is individual and independent, as is the accountability for decisions made. Decision-making responsibility and accountability are interdependent and relational in the overlapping sphere.The first of the following excerpts illustrates autonomous action in the nursing-unique sphere; the second exemplifies autonomous decision making and action in the overlap sphere of practice.Although U/O spheres of practice and the corresponding types of decision making have long been heralded in the literature,24,27–30 only recently has adoption of the concept in nurses' professional practice become evident. In a 1980 classic book on autonomy, Mundinger28 emphasized that autonomous practice is not a nurse providing medical care without medical supervision or a nurse practicing medicine without a license; rather it is a nurse providing nursing therapy that complements and at times overlaps medical therapy. In the same year, the American Nurses Association and the American Medical Association combined commission on collaborative practice recognized the U/O spheres of practice and called for "the formal development of Scope of Practice documents."29 In their 1980 Social Policy Statement, the American Nurses Association emphasized that clinical autonomy and collaborative relationships between nurses and physicians are true partnerships in which power is held and valued by both participants with recognition and acceptance of separate and combined spheres of activity, responsibility, and accountability.30 In the 1983 original magnet hospital study,31 all of the autonomy examples cited were in the nursing-unique sphere of practice, for example, preadmission programs for children, counseling for pregnant adolescents, and support services for senior citizens. The overlap sphere was alluded to when it was noted that in settings without a house staff, nurses perceive themselves as having greater autonomy and responsibility in decision making and in management of patient care. The current magnet program of the American Nurses Credentialing Center defines autonomy as "independent judgment exercised within a multidisciplinary approach to patient care."32 This somewhat contradictory definition does not address U/O spheres of practice or corresponding types of decision making, nor are these addressed in either the Forces of Magnetism or in the suggested sources of evidence.32Despite the overt approval by national professional nursing and medical organizations, no evidence of widespread, consistent use of the concept of U/O spheres of practice was reported until 2001, when staff nurses in 14 magnet hospitals were interviewed.4 When asked to describe a situation in which they practiced autonomously, many nurses inquired, "Do you want me to describe a patient care or a nursing care decision?" Interviewees' descriptions indicated that autonomous patient care actions were decisions that "extend beyond the usual parameters of nursing to other disciplines,"2,4 essentially the same as decisions made in the overlapping sphere of practice. Nursing care decisions were independent actions focused in the nursing arena only, the same as the nursing-unique sphere of practice. Scope of practice emerged as 1 of the 3 dominant themes when the examples and descriptions of autonomy were categorically analyzed; frequency of action and organizational sanction for autonomous practice were the other 2. A total of 43% (n=117) of the nurses cited autonomous patient care actions; 17% (n=47) cited nursing care actions.2Tracking the performance of several research samples (Table 1) on the U/O item of the Autonomy sub-scale of the EOM provides further evidence of the power and use of this U/O spheres concept as a critical element in autonomous practice in magnet work environments. The Autonomy subscale of the EOM II contains this item: "On this unit, nurses make independent decisions within the practice sphere of nursing and interdependent decisions in those spheres of practice where nursing overlaps with other disciplines such as medicine and respiratory therapy." Not only do nurses in magnet hospitals consistently score significantly higher on the total Autonomy scale than do their counterparts in nonmagnet hospitals,8,17 but the percentage of staff nurses in magnet hospitals who responded affirmatively to the U/O item increased from 83% of 279 nurses in 200133 to 90.2% of 3602 nurses in the 2003 study8 to 94% of 3510 nurses in the 2006 autonomy structure-identification study.12,13 The percentage of positive responses in comparison hospitals has remained low, decreasing from 38% in 2003 to 29% in 2006.8,17The results of the autonomy structure-identification study12,13 provide additional evidence of increased recognition and use of the concept of U/O spheres of practice. When presented with 8 of the 34 definitions of autonomy found in the literature, 60% of the 267 staff nurses, nurse managers, and physicians interviewed selected 1 of the following 2 definitions as representative of their understanding of the concept of autonomy.12 Both definitions recognize the concept of U/O spheres of practice.A total of 31% of the 267 interviewees selected 1 of the 2 following literature definitions to represent their understanding of autonomy. Both of these definitions acknowledge mainly or only the overlap sphere.The U/O concept and corresponding different types of decision making are not included in the American Nursing Credentialing Center definition of autonomy for magnet designation.32 Not all magnet hospitals score at the level noted in the previous samples, nor do all nonmagnet hospitals score low. The percentage of nurses in magnet hospitals who respond affirmatively to the item on U/O spheres of practice is sometimes as low as 37%, and in some comparison hospitals, as many as 85% of nurses respond affirmatively to this item.The differences in scores occur because in some hospitals, approval and sanction from the leadership create the interest, excitement, and security for clinical nurses to practice autonomously. One staff nurse remarked as follows:Several nurses from a single unit who came to the 2006 interviews as a group had this to say:Executive-level nurses, managers, and staff nurses in the 8 magnet hospitals that participated in the autonomy structure-identification study13 provided considerable evidence of the support present in their hospitals for autonomous practice. The following are examples:If the risk involved is so great, why should staff nurses make autonomous decisions? Such decisions are essential for patients' safety, nurses' job satisfaction, and nurse retention.2,3,12,19,28,33 Nurses want to make decisions that promote quality patient care. They perceive autonomy as an, if not the, essential component of professional practice.12,15 Physicians rated autonomous decision making as the highest indication of competent performance of staff nurses.1 Most recently, the Institute of Medicine34 recommended that a higher level of clinical autonomy be given to staff nurses and that they be trusted and supported in using the outcomes of evidence-based practice initiatives to make decisions about patients' care.Unmet patient needs and rapidly changing conditions are the stimuli for autonomous actions. In the autonomy structure-identification study,12,13 we inquired, "What situations or patient needs most frequently motivate you (or staff nurses) to autonomous action?" Interviewees described examples that fit into 6 domains. Table 2 presents each of these domains with its primary motivating factor or patient need. Examples of the emergency domain were often in the overlap or combined sphere of practice. Need to rescue and patient advocacy tied for highest frequency, with physicians citing more need to rescue and nurses more patient advocacy examples.12,13Triage decisions varied by unit and tended to be unit specific; these included referrals to the proper place and person for care and the adequacy of preparation for treatments, procedures, or surgery. Patients with multiple comorbid conditions who had several physicians ordering multiple and sometimes conflicting treatments gave rise to the coordination and integration domain.12,13These autonomy domains are not pure; overlap occurs. Although all domains can be present on any clinical unit, patients' needs tend to vary by units. The emergency and need to rescue domains are far more common in critical care units than in other units. Coordination and integration are predominate needs of patients in geriatric and rehabilitation units but are also needed, perhaps to a lesser degree, in critical care. Each domain requires different knowledge, assessment, competencies, and decision-making skills.12,13 Most likely autonomous practice would increase, along with improved quality of patient care and nurses' job satisfaction, if unit orientations and meetings included discussions of the frequently made autonomous decisions on the unit and the knowledge and competencies needed to make those kinds of decisions.On which clinical units do nurses report the most autonomous practice? In the autonomy structure-identification study,12,13 we selected our interviewees from those clinical units in the 8 magnet hospitals in which staff nurses had previously confirmed that they could and did function autonomously. Every oncology unit (n=18) in the 8 hospitals scored high enough on the Autonomy subscale of the EOM to be selected for individual interviews with staff nurses, managers, and physicians. All of the orthopedic units had similarly high scores, but only 7 of the 8 hospitals had such units and then usually only 1 unit. Procedural outpatient clinics had the next highest scores. Intensive care units (ICUs) were markedly underrepresented in the high-scoring unit sample; only 8 of 34 units in the 8 hospitals reported extensive autonomous practice as measured by the Autonomy score on the EOM.12,13Data from the autonomy structure-identification study12 is not the only evidence for this ordering of units with respect to degree of reported autonomous practice. Almost the same unit lineup was found in the 2006 six essentials structure-identification study1,14–16 on 101 units in 8 other magnet hospitals. Oncology units reported the highest scores for autonomous practice; next, in order, were procedural out-patient departments and orthopedic units; combined ICUs were third from the bottom.35The placement of ICUs in the unit lineup is puzzling and counterintuitive, particularly because ICU nurses usually report a greater need for and score higher in autonomy than do nurses in other specialty areas.36 Why did nurses in some ICUs (8 of 34) report appreciably more autonomy than did nurses in other ICUs? To explore this question, we conducted informal inquiries of individual and groups of ICU nurses during the 2006 American Association of Critical-Care Nurses National Training Institute in Anaheim. We explained the counterintuitive finding and asked the nurses for their ideas, insights, and explanations as to why this finding might have occurred. A dominant explanation was that the increased use of intensivists and having physicians present almost all the time preclude the need for autonomous decision making in the overlap area. Another popular explanation was that autonomy in the nursing-unique sphere was so much a part of everyday practice in the ICU that ICU nurses did not consider it or report it as autonomy. In a completely different study, Canadian nurses reported a similar explanation for why ICU nurses may have low autonomy scores.25 The following incident related to us by an experienced ICU nurse, describes what one nurse learned from a colleague about the importance of autonomous, "caring" decisions and actions in the nursing-unique sphere of practice.The 267 staff nurses, nurse managers, and physicians working on the 74 units in 8 magnet hospitals who had high scores in autonomy in the autonomy structure-identification study,12,13 described structures and best practices that enabled staff nurses to make autonomous decisions. Three structures and best practices that have particular relevance for clinical nurses are discussed in the following sections.Autonomy cannot be safely and effectively pursued with appropriate accountability without an understanding of the U/O spheres of practice by staff nurses, management, administration, and other disciplines. This pursuit of autonomy cannot be done without agreement about what clinical autonomy is and how it is to be interpreted, at least on each unit and in each hospital. Staff nurses can exert leadership in this endeavor. They can bring the issue to a staff meeting. They can ask their colleagues to describe their understanding of the concept of autonomy. They can discuss the risks and benefits involved. The 4 definitions provided earlier may be a good place to start. Staff nurses can discuss the differences between independent decision making and interdependent decision making that involves input from other disciplines. What can and should nurses do if their assessments and observations indicate that patients' needs are not covered by current orders or require action outside nurses' current scope of practice? As a physician said, "Pretending that risks and impediments to autonomous practice don't exist will not make them go away."As professionals, clinical nurses, particularly those in magnet hospitals, have the obligation and responsibility "to do something about the autonomy mess" not only in the places where they work but also with their school alumni and faculty, with the American Nurses Credentialing Center, and with professional nursing specialty organizations. Nurses can ask themselves, what were they taught about clinical autonomy in their nursing programs? What have they learned since beginning professional practice, and what would they like to see incorporated into the curriculum in the schools of nursing they attended? The original Forces of Magnetism originated from interviews with staff nurses and directors of nursing in the original 41 magnet hospitals. Who better than professional nurses in currently designated magnet hospitals to advise the American Nurses Credentialing Center about the current practice of clinical autonomy and evidence to support that practice?32 The American Association of Critical-Care Nurses is the only professional specialty organization that recognizes U/O spheres of practice as essential for autonomous practice.37 If more impediments to autonomous practice are perceived by critical care nurses than by nurses on other units, critical care nurses need to initiate discussion of this issue, not only in ICUs and in the employing hospitals but also in meetings of their specialty organization.Since the 1970s, scope of practice has been termed redefining domain boundaries, shifting limits on action, knowledge and responsibility to meet patient needs, performance of skills beyond professional jurisdiction, role enlargement or expansion, and situational credentialing.13 Negotiation relative to scope of practice is particularly necessary in the overlapping sphere of practice. Interviewees in the autonomy structure-identification study12,13 identified 3 levels or patterns of autonomous decision making and action. The first, do and inform later, is based on a history and mutual trust between physicians and nurses and is probably the most frequent autonomous action in the overlapping sphere of practice.The second level of autonomous action, persist until the patient gets what he or she needs, involves repeatedly contacting the same physician, contacting other physicians, going "up the chain of command," or electing not to follow an order or a protocol that the nurse judges to be inappropriate for this patient in this situation. The second excerpt about the elderly patient with Alzheimer's disease in the section "Crucial Components" is an example of this type of autonomous action. The following is another.True interdependent decision making characterizes the third level of autonomous practice. The following excerpt from an ICU nurse in an autonomy workshop illustrates this level.Renegotiation of scope of practice also takes place in unit operations meetings or as a part of interdisciplinary rounds, such as are frequent in medical ICUs, neonatal ICUs, stroke units, and trauma units, or in regularly scheduled interdisciplinary meetings to develop, assess, and evaluate nurse order sets, collaborative practice orders, critical pathways, protocols, and rapid response team directives. It is through the process of developing, assessing, and evaluating the effectiveness of orders and protocols that renegotiation of what decisions and activities are best for quality patient care takes place. Interviewees, both nurses and physicians, were concerned about "one size fits all" protocols with required implementation, such as restraints or electrolyte replacement, for all patients in a specific group. These protocols were strongly criticized as constrictive, restrictive, and of little value in autonomous decision making because they provided little room for judgment and appropriateness for individual patients. Sometimes a protocol does not fit this particular patient, at this particular time, in this particular context.The role of clinical nurses in all of this negotiation is to participate, lobby for clarification, and identify what is needed. In the nursing-unique sphere of practice, renegotiating scope of practice is dictated by best practices, new developments in the field, or new technologies and is largely a matter of intraprofessional dialogue. The overlap sphere requires interprofessional dialogue and negotiation.In the 2006 autonomy structure-identification study,12,13 interviewees identified administrative sanction, approval, and support as a best practice enabling staff nurses to make autonomous decisions. This finding is not new. This same best practice was 1 of the 3 themes that emerged from the analysis of the hundreds of examples and descriptions of autonomous practice provided by the 279 staff nurses interviewed in 2001.2 "I can not and will not practice autonomously if I don't have at least the support and approval of my nurse manager." Interviewees noted that if a nurse did not have this support, the only alternative was do it "back stage," or as their Canadian counterparts said, "You've got to go in the 'back door'25 and live with the risk that you may get jumped on for trying to do the best you can for your patient." Several interviewees remarked that the only support they got for autonomous decision making was from their peers or from physicians.13Additional autonomy-enabling structures and best practices were identified by interviewees or through an in-depth analysis of operational and evaluation data collected from each of the 8 hospitals as part of the autonomy structure-identification study,12,13 The most extensively cited structures were evidence-based practice teams, activities, and initiatives.13 To practice autonomously, a nurse must know, must have up-to-date knowledge backed by research and evidence: "Our evidence-based practice teams generate the knowledge upon which protocols and critical pathways are based." The most effective teams were interdisciplinary evidence-based practice teams of nurses, physicians, pharmacists, and therapists organized around patient care problems.13A second autonomy-enabling practice signifying administrative sanction was the inclusion of autonomy-related concepts in the criteria for career ladder programs or in the criteria for a similar bonus value-added program.13 All 8 hospitals in the autonomy structure-identification study12,13 had some type of professional advancement program. Often, the criteria for advancement centered on participation or leadership in educational or research activities or on council or practice improvement agendas. Of the 8 hospitals that had the highest autonomy scores of more 100 hospitals tested with the EOM, 2 based all or a large part of their career ladder criteria on the steps and components of autonomy, specifically designating the type of decision making, sphere of practice, and risk taking as advancement criteria.13 In 6 of the 8 hospitals, spheres of practice and related decision making were spelled out in documents entitled Scope of Practice, Definitions of Nursing, and Models of Professional Practice.13A third structure that communicates administrative/organizational sanction and support for autonomy is performance appraisal documents. Four of the 8 hospitals cited this type of sanction.13 In 3 of the 4 hospitals, identification of practice spheres and types of decision making were listed as subbehaviors and competencies; in 2, "negotiating scope of practice with physicians and other disciplines" was cited as an advanced staff nurse competency; in all 4, independent and interdependent decision making were defined and listed as major competencies of a professional nurse. The fact that only 4 of these 8 magnet hospitals cited sphere of practice and type of decision making in their performance appraisals, and that in none of the 8 hospitals were these competencies expected of entry-level nurses, supports staff nurses' contentions that autonomous decision making is exceptional rather than beginning or standard nursing practice.13Staff nurses can take action to obtain the administrative support needed for autonomous practice. Nurses must ask for the support; they have an obligation to make their expectations known.14,15 If a nurse wants to practice autonomously and feel that patients need and deserve the best that the nurse can give, he or she must ask that clinical autonomy be defined, discussed, and included in performance appraisals and in clinical advancement criteria. Nurses must bring the issue to practice councils.Sometimes nurses are made to feel that they do not have the support of leadership and/or management in making autonomous decisions in the best interests of patients. An experienced nurse related the following:Unless these issues are openly discussed and brought to the attention of management and leadership, clinical nurses will never really find out if they are or are not expected to function autonomously.Staff nurses in magnet hospitals describe autonomy as the freedom to act in the best interests of patients, to make independent decisions in the nursing sphere of practice and interdependent decisions in those spheres in which other disciplines overlap with nursing. Autonomous practice is essential for safe and quality patient care and for nurses' job satisfaction. Nurses on oncology, procedural outpatient, and orthopedic units report more autonomous practice than do nurses on other units. Six domains of autonomy based on patients' needs were described; many of these are unit specific. Improving the work environment so that nurses can function autonomously is a responsibility of both leadership and clinical nurses. Clinical nurses can exert leadership by defining autonomy, at least for their unit or service. Periodically renegotiating the scope of practice and obtaining administrative support and sanction will do much to secure autonomy-friendly work environments.