Types of Intensive Care Units With the Healthiest, Most Productive Work Environments

Claudia Schmalenberg, RN, MSN
Marlene Kramer, RN, PhD

The American Association of Critical-Care Nurses (AACN) defines a healthy work environment as a work setting in which structures are designed so that nurses can achieve 2 outcomes: meet organizational objectives and achieve personal satisfaction in their work.¹ AACN has identified 6 standards or relationship-centered principles of professional performance² through which these outcomes are to be achieved. Environment is the aggregate of conditions and circumstances that influence an organism, so each of the 6 standards is essential to a healthy work environment. The standards are interdependent; none can be considered optional.³

Intensive care units (ICUs) have staffing and other structures that differ from those of other clinical units. These structures differentially affect functional care processes and relationships that, in turn, affect outcomes such as nurses’ job satisfaction and their ability to give quality care to patients. Differences among types of ICU units—adult and pediatric, medical and surgical—also have been noted.^4-6 Combining samples of nurses from various categories of ICUs may mask differences in structures that enable care processes.

The Essentials of Magnetism (EOM) is a psychometrically sound instrument³ that measures 8 functional processes essential to a productive work environment. The 8 processes are highly intercorrelated and interdependent; all are essential to a healthy work environment. The AACN standards and the EOM are not identical. The standards were identified by leaders, experts, and a professional organization; the EOM was compiled from the perspective of staff nurses working in magnet hospitals.⁷ Both the standards and the EOM focus on processes or relationships, and both emphasize that it is not any one process or relationship but the aggregate that constitutes a productive, healthy work environment. This congruence and alignment between the standards and the EOM are sufficient to make the EOM a suitable instrument for answering the questions that guided our study: How healthy are ICU work environments? Do some types of ICUs report healthier work environments than do others?

Background

Structures

The structural elements and attributes of ICUs that are linked to a healthy practice environment are a physical layout that allows constant observation and immediate access to patients; a high level of rapidly developing technology; competent, experienced nurses; a low nurse to patient ratio⁸; longevity of contact between nurses and physicians^9-11; and a high degree of medical specialization.^10-12 ICUs also have high “medical pervasiveness,” that is, a relatively small number of physicians who are called and who visit the unit frequently and for longer periods than do physicians in other units.¹³ Bedside rounds with physicians, nurses, healthcare workers from other disciplines, the patient, and the patient’s family all discussing the patient’s progress and daily and long-term goals are characteristic of ICUs, particularly medical ICUs.^10,11

Processes and Outcomes

The 6 relationship processes identified in the AACN standards are skilled communication, true collaboration, effective decision making, staffing that matches patients’ needs and nurses’ competencies, meaningful recognition, and authentic leadership. Some processes such as effective decision making have been positively linked to ICU structures,^13,14 and ICU structures have been linked to patient outcomes such as mortality and to nurse outcomes such as burnout, job satisfaction, stress, and turnover.^13,15 The functional processes and relationships that constitute a productive, healthy work environment have not been measured and studied in their aggregate.

The results of empirical studies of the effects of ICU structures on processes and outcomes have been mixed. In one study,¹⁴ ICU nurses had a greater need for autonomy and scored higher in autonomy than did nurses in other types of clinical units. In another study,¹⁶ researchers found no differences in autonomy scores between nurses in ICUs and emergency departments and nurses in general medical-surgical units. In still another study,⁴ ICU nurses scored the lowest of all groups in autonomy.

Structures and nurse outcomes often are linked by comparing scores of nurses from ICUs with scores of nurses from other units. ICU nurses are reported to have more occupational stress, less job satisfaction, and greater turnover than are nurses in other types of units,¹³ although it also is reported that medical-surgical nurses have higher occupational stress and turnover than do nurses in other units.¹⁶

In a study of 55,516 registered nurses (2900 work groups) in 206 hospitals in 44 states, Boyle et al¹³ reported that work group satisfaction was moderate across 10 types of clinical units. Nurses in pediatric units were the most satisfied of all, those in emergency departments and perioperative services were the least satisfied of all, and ICU nurses were the most satisfied of nurses in the 7 remaining types of units.¹³ When different clinical units were compared with respect to 8 attributes essential to a productive work environment, ICU nurses scored higher on collegial/collaborative relationships between nurses and physicians and perception of adequate staffing and lower on nurse manager support than did nurses from other units (C.E.S. and M.K., unpublished data, 2007).

In a classic study that showed linkages between structures, processes, and outcomes in assessing the quality of healthcare and work environments, Knaus et al¹⁵ examined the relationship between ICU structures and the patient outcome “mortality less than would be expected by acuity.” The findings indicated that the significantly lower mortality rates in the ICUs studied nationwide were due not to the structural elements of ICUs but rather to the processes of teamwork and collaboration between nurses and physicians. These results illustrate the fundamental principle that, although structure is critically important, structure alone does not produce outcomes. Structures enable processes that lead to outcomes.¹⁷

Types of ICUs

In 3 studies,^6,18,19 nurses in medical ICUs (MICUs) reported more favorable components in their work environments than did nurses in other types of ICUs. In a study of 2323 nurses in 110 ICUs in 64 hospitals, Cimiotti et al⁶ found that nurses in MICUs and medical-surgical ICUs (MSICUs) perceived higher staffing levels than did nurses in coronary care units and surgical ICUs (SICUs). The degree of collaboration between physicians and nurses as perceived by nurses was related to positive outcomes for patients in MICUs but not in SICUs or MSICUs. Baggs et al¹⁸ reported that the degree of collaboration as perceived by physicians was not associated with outcomes in any type of ICU. In a study by Ferrand et al¹⁹ of 3156 nurses and 521 physicians from 133 French ICUs (90 MSICUs, 22 SICUs, and 21 MICUs), MICU nurses believed they were more involved and more satisfied with “end-of-life” care decisions than were nurses in SICUs or MSICUs.

The aggregation of ICUs into different types was not consistent across these studies,^6,18,19 making the results difficult to interpret. In some, coronary care units were grouped with MICUs; in others, coronary care units were studied as separate types of ICUs. None included neonatal ICUs (NICUs).

Measuring Nurses’ Work Environments

Few tools are available to measure nurses’ work environments. Studies^20,21 in which environments were measured by using conceptually derived subscales from the Nursing Work Index²² were based on individual rather than unit level data, lacked a theoretical base, and measured “presence” of the attribute without regard to the steps or components of the process or to the respondent’s definition of the concept. For example, compare the statement “I can practice autonomously” with “Nurses on this unit make independent care decisions in that sphere of practice that is uniquely nursing.” Such differences make it difficult to relate, compare, or interpret the results.

With the EOM, both the components of the work environment and the composite work environment can be measured; 90% of the items are written from a clinical unit perspective and the remaining 10% are organizationally and unit based.³ The EOM has a long developmental history. In 1984, 65 characteristics of a magnetic work environment, confirmed by the original investigators, were abstracted from the original magnet hospital report, and a tool to measure job satisfaction and productivity was developed.²³ After administration to thousands of nurses during a 12-year period, the tool was condensed to the 37 most frequently selected items. In 2001, staff nurses in 14 magnet hospitals were asked to identify the 10 attributes most important to “being able to give quality patient care” (productivity).⁷ In the magnet hospital study,²⁴ 4 outcome criteria—attraction, retention, productivity, and job satisfaction—were used to designate magnet or excellent work environments. In a causal modeling study,²⁵ productivity accounted for more than 80% of the variance in job satisfaction, attraction, and retention.

Hence, in the 2001 study,⁷ staff nurses were asked to select the essential environmental attributes on the basis of productivity alone. The 8 attributes identified by staff nurses in magnet hospitals were as follows:

1. Working with clinically competent peers
2. Collegial/collaborative relationships between nurses and physicians
3. Clinical autonomy
4. Nurse manager support
5. Control over nursing practice
6. Perception that staffing is adequate
7. Support for education
8. A culture in which concern for the patient is paramount (values)

On the basis of interviews with 279 staff nurses and 132 leaders and managers in 14 magnet hospitals⁷ and participant observations of nurses in 12 other magnet hospitals,²⁶ grounded theories were generated.^3,12,27 Items to measure each attribute of a productive work environment were developed on the basis of these theories and the definitions and descriptions provided by nurses during interviews. Each attribute is measured by using a subscale.

Weighting studies were done to determine relative importance of steps and components of the process.³ For example, physician-nurse relationships based on mutual power, trust, and respect (collaborative) are more instrumental in enabling quality patient care than are student-teacher or “friendly stranger” relationships, hence the item has greater weight in the scoring. The weighted, composite score for the 8 relationships or processes is a measure of a healthy, productive work environment; it is labeled “professional job satisfaction” to signify that it is job satisfaction due to professional productivity. Staff nurses describe this variable as an environment “that helps me do a good job,” “in which I can make a difference in the care patients receive,” or where “what I do helps patients get better and stay healthy.”

Alignment is considerable between components of a productive work environment as measured by the EOM and the 6 AACN standards. Both the EOM and the standards are based on relationships or processes. The skilled communication standard is most closely related to the process of “working with other nurses who are clinical competent” but also to “support for education.” An almost direct parallel exists between the true collaboration standard and the process of establishing collegial and collaborative relationships between nurses and physicians. True collaboration also refers to one of the steps in the clinical autonomy process, “respect for the unique knowledge and ability of each profession.”^2(p190) Effective decision making is related to both the “clinical autonomy” and the “control of nursing practice” processes of the EOM. Appropriate staffing parallels the EOM process of “perceived adequacy of staffing.” The meaningful recognition standard is most closely related to “control of nursing practice,” and the authentic leadership standard is related to “nurse manager support.”² Both the EOM and the AACN standards recognize the interrelationship and interdependence of the components whose aggregate constitutes a snapshot of a specific environment. The EOM is used to measure a productive work environment; the AACN standards define healthy as productive and satisfying.²

Objectives of Study

The purpose of our study was to answer the following questions: To what extent do ICU nurses confirm a healthy work environment? Are there differences in perception by type of ICU?

If some types of ICUs excel, systematic study of units that report healthy work environments will permit identification of structures and practices that, when implemented, would improve the practice environment of other clinical units. Analysis of the individual processes and relationships that lead to productive work environments will enable assessment of the impact that the AACN standards have had on improving the work environment of nurses in ICUs and will suggest specific areas and strategies for change and improvement.

Design and Sample

A cross-sectional descriptive design with strategic sampling was used in this secondary analysis of data from a larger study²⁸ designed to identify organizational structures and practices that enable processes and relationships essential to a productive work environment. The complete sample consisted of 2990 staff nurses from 206 clinical units in 8 magnet hospitals. The ICU subsample was 698 staff nurses from 34 ICUs grouped into 4 types: (1) medical and coronary care units labeled MICU; (2) surgical, cardiovascular, and trauma units labeled SICU; (3) neonatal and pediatric units labeled NICU; and (4) mixed medical-surgical critical care units labeled MSICU. The magnet hospitals selected for the strategic sample had the highest or second highest composite EOM score in the 8 regions of the country; selection between highest and second highest was done to balance the academic and community hospital samples.

Method

Instrument

The EOM was used to measure a healthy work environment as defined by AACN. For 7 of the subscales, participants respond to a 4-point Likert scale ranging from strongly agree to agree to disagree to strongly disagree. For the subscale on the relationships between nurses and physicians, the options are as follows: true for most physicians, most of the time; true for some physicians, some of the time; true for 1 or 2 physicians on occasion; not true for any physicians. Some items are reverse scored. The sum of the weighted items equals the score for the subscale. Professional job satisfaction, equivalent to productive work environment, is the composite score for the 8 subscales. Two global-item outcome indicators were used to measure overall job satisfaction and nurse-assessed quality of care. Both are 1 to 10 scales (10 high), and benchmarks are provided. Content validity indices for the 8 subscales range from 0.88 to 1.00, with a median of 0.92.³ Cronbach α’s for subscales and outcome measures range from .80 to .90, with a median of .88.³

Procedure

The EOM was administered to the staff nurse population in each hospital during a 6-month period in late 2005 and early 2006. Because the study was one of work environment, only clinical units with a complement of more than 5 registered nurses (to protect anonymity) and a response rate of 50% or more (to ensure representativeness) were included. After approval was obtained from the institutional review board, EOM data were collected by on-site investigators.

Data Analysis

Univariate analysis of scores on the EOM subscales, EOM total score, and outcome measures by experience, education, certification, type of hospital, and ICU subtype was used to detect significant differences. Multivariate analysis procedures were used to control for differences.

Results

Description of the Sample

A total of 66% of the ICU nurses had a baccalaureate or higher degree; SICUs had the largest percentage (71%) of nurses with a baccalaureate degree or higher (Table 1). Mean years of experience ranged from 12 in MICUs and SICUs to 14 in MSICUs, with a mean of 13. Among the ICU nurses in the sample, 27% were nationally certified; 60% had earned the CCRN certification; 23%, the RN, C certification; and the remaining 17% had certifications scattered among 15 different specialties. The MSICUs had the highest percentage (35%) of nationally certified nurses. None of the differences among different types of ICUs with respect to experience (x² = 25.8; P = .10), education (x² = 16.3; P = .18), or certification (x² = 5.0; P = .17) were significant. Types of ICUs differed significantly between the 3 academic hospitals and the 5 community hospitals (x² = 163.9; P < .001). A total of 91% of the MSICUs were in community hospitals; 74% of the NICUs were in academic hospitals. SICU and MICUs were evenly distributed between the 2 types of hospitals.

Which Types of ICUs Report the Healthiest Work Environments?

In this strategic sample of 8 magnet hospitals scoring above the mean²⁹ on the National Magnet Hospital Profile, the sample of 698 ICU nurses reported a mean score of 292 for professional job satisfaction, a score of 7.18 for overall job satisfaction (10-point scale, 10 high), and a score of 8.31 for nurse-assessed quality of care.

Covariate analysis of the EOM process and the outcome scores, with education, experience, certification, and type of hospital controlled for, revealed no significant differences among the 4 types of ICUs in terms of education or certification. Differences in scores by experience and type of hospital were significant (Table 2).

Follow-up analysis (data not shown) indicated that nurses with 3 years or less of experience and those with more than 30 years of experience scored significantly higher (P = .001) on the essentials, support for education and patient-centered values, and on the outcome, overall job satisfaction, than did the other groups. For nurse-physician relationships, nurses with more than 20 years of experience had significantly higher scores than did nurses with more than 3 and up to 5 years of experience. The consistently lowest scoring groups were the nurses with more than 3 and up to 5 years of experience, more than 10 and up to 15 years of experience, and more than 5 and up to 10 years of experience. The 360 nurses from community hospitals scored significantly higher on the essentials, clinically competent peers (P = .001), control over practice (P = .003), and adequacy of staffing (P = .002), and on the outcomes, professional job satisfaction (P = .04) and quality of care (P = .007), than did the 338 nurses from academic hospitals. Nurses in academic hospitals scored higher on nurse manager support (P = .001) and patient-centered values (P = .007).

Covariate analysis indicated that the primary differences in EOM process and the outcome scores were due to the type of ICU (Table 2). Post hoc analysis indicated that NICUs had the highest scores of all units on the outcome variables: professional job satisfaction, overall job satisfaction, and nurse-assessed quality of care. Nurses in NICUs scored significantly higher than did nurses in MSICUs on professional job satisfaction, the overall measure of a healthy work environment used in this study. Mean scores and source of significant differences are presented in Table 3. Nurses in NICUs scored significantly higher than did those in SICUs on nurse-assessed quality of care. Nurses in NICUs had higher scores than did nurses in the other types of ICUs on the components of a healthy work environment, particularly the nurse-physician relationship, control of nursing practice, perceived adequacy of staffing, and patient-centered values. MICUs scored higher than all other types of ICUs and significantly higher than MSICUs on the other 4 components of a healthy work environment: support for education, nurse manager support, clinical autonomy, and clinically competent peers.

Item analysis was done to ascertain steps and components of the EOM processes that accounted for significant differences in subscale scores. We used the percentage of nurses responding affirmatively (strongly agree and agree) rather than mean item scores because the percentages seemed conceptually more meaningful. Table 4 shows those items for which differences were significant. NICU nurses reported the highest percentage of positive factors in their work environments such as equal trust, power, respectful working relationships with physicians, and cohesive work groups, and the absence of negative factors such as bureaucratic rules that inhibit decision making and a hospital culture that is reluctant to try new things.

Conclusions and Implications

Two-thirds of the ICU nurses in this sample had a baccalaureate or higher degree. This percentage is well above the 50% goal set by nurse executives in community hospitals and close to the 70% goal set in academic hospitals.³⁰ These ICU nurses were quite experienced, with a mean of 13 years of experience. About one-quarter were nationally certified. This percentage (27%) is virtually the same as that reported in a recent study (C.E.S. and M.K., unpublished data, 2007) of 10 514 nurses in 18 magnet and 16 comparison hospitals, although ICU nurses were the clinical group with the most national certifications. This percentage of certified nurses seems quite low in our strategic sample.

Nurses in the ICUs represented in this study reported highly productive work environments that reflect the 6 standards fashioned by AACN as measured by the professional job satisfaction score of the EOM. This score for the ICU nurses in this study was 292, a mean score that exceeds both the 2004 National Magnet Hospital Profile mean score of 289 and the mean score of 287 reported for 18 magnet hospitals in recent studies²⁹ (C.E.S. and M.K., unpublished data, 2007). As of the end of 2006, 76 magnet hospitals had been tested with the EOM; the 95th percentile for professional job satisfaction for all magnet hospitals is 291.

When asked how satisfied they were with their current nursing job, considering all aspects of the job, salary, and fringe benefits as well as values, ideals, and goals, ICU nurses rated their overall job satisfaction as 7.18 on a 10-point scale, where 10 is the highest score. This rating is within range but below the National Magnet Hospital Profile mean score of 7.74, yet it is higher than the 6.86 mean score reported for other magnet hospitals.²⁹ The 7.18 score is slightly above the 95th percentile score of 7.09.

The mean score for nurse-assessed quality of patient care of 8.31 was one of the highest scores obtained in any study that used the EOM. In both the study of 18 magnet hospitals (C.E.S. and M.K., unpublished data, 2007) and the National Magnet Hospital Profile,³⁰ the mean score was 8.04. The 95th percentile score was 8.26. Thus, the ICU nurses in our study not only confirmed healthy work environments but rated their overall job satisfaction as high and rated the quality of care they give to patients as outstanding.

Relationships between education and experience and outcomes such as healthy, productive work environments, overall job satisfaction, and nurse-assessed quality of care must be empirically determined. Our results indicate that baccalaureate-educated nurses are well prepared to avail themselves of opportunities that enable them to engage in processes and relationships that lead to job satisfaction and quality patient care, perhaps better prepared than are their less educated counterparts. Years of experience, on the other hand, do not seem to progress in a tidy fashion; both very inexperienced nurses and very experienced nurses confirmed healthier work environments than did other groups with different levels of experience. Perhaps some nurses have 13 years of experience whereas others nurses have 1 year repeated 13 times. Is it possible to coach/mentor nurses and plan work activities so that each year employed is a year of high-quality experience?

What part does specialty certification play in increasing nurses’ ability to use structures and opportunities presented to improve satisfaction, productivity, or processes such as autonomy and clinical competence? Published reports suggest positive relationships between certification and clinical competence³¹ and between certification and empowerment.³² However, a national critical care survey indicated a perceived lack of organizational support for specialty certification.³³ In our study, MSICUs had the largest percentage (35.2%) of certified nurses. But these nurses scored lowest on the clinically competent peers essential, and item analysis indicated the lowest perception on the item “certification is a mark of clinical competence.”

This result seems to indicate that specialty certification was not highly valued by MSICU nurses or by other types of ICUs. Either nurses do not recognize the potential of specialty certification as a baseline for many of the processes and relationships inherent in a healthy and productive work environment, or financial support and recognition are insufficient, or managers may need to refocus on other types of educational programs to serve as building blocks for certification. The value and relevance of national certification as it affects a healthy work environment must be empirically studied. Without additional organizational support and financial incentives, specialty certification may remain at a relatively low level.

NICU nurses in this study scored the highest on 4 of the process variables as well as professional and overall job satisfaction and nurse-assessed quality of care; MSICU nurses scored the lowest on most variables. What is it about the NICU work environment that leads clinical nurses to proclaim the NICU as such an ideal work environment? It is probably not nurse attributes of education, experience, or certification because we found no differences in these demographic characteristics by type of ICU.

The differences between different types of ICUs may be due to the structural feature degree of specialization. Empirical studies^10,11,34 indicate that a high degree of specialization is directly linked to development of collegial and collaborative relationships between nurses and physicians, enactment of clinical autonomous decisions, and perception of high clinical competence. Of the 4 types of ICUs, NICUs have both age specialization and medical specialization. Patients in NICUs are usually under the supervision of a relatively small group of neonatologists. This arrangement leads to more frequent contact between the same physicians and nurses and hence more opportunity to develop collaborative and collegial relationships.^10,11

The greater, deeper, more consistent family involvement characteristic of NICUs may account for the high professional and overall job satisfaction scores. MSICUs are the least specialized, being the equivalent of medical-surgical units in the intensive or critical care area. In contrasting NICUs and MSICUs, the age of the patient also may affect degree of specialization, because adult ICU patients often have comorbid conditions that require a breadth of knowledge, skill, and competence among the nurses, whereas NICU nurses need depth of knowledge and competence for a narrow age range.

All of the ICUs in this study, but particularly the NICUs, are models of healthy, productive, professional work environments for nurses. Much can be learned from studying their structures, practices, and features. Just as none of the AACN 6 standards of a healthy work environment is optional, neither are the functional processes that staff nurses identify as constituting a productive work environment. The 8 essentials are intercorrelated and interdependent, some to a greater degree than others. Although all these factors contribute to a healthy work environment, comparison of the performance of a unit group of nurses on individual processes with the high standards set by the nurses in this study will yield information and direction that will permit formulation of strategies to improve specific nursing work environments.

Relationships among functional processes (essentials) also must be studied. Competence is the basis for the mutual or equal power, trust, and respect that characterize collaboration between physicians and nurses.^10,11 Physicians want nurses to function autonomously, particularly in “need to rescue” situations, but only if the nurses are competent.^12,32 The major goal of the essential “support for education” is to advance clinical competence. A key support behavior of nurse manager support is “making it possible for staff to attend seminars, programs, and other educational activities.”³⁵

Specific help to individual units and hospitals in creating and sustaining healthy work environments can be further facilitated through analysis of items on each of the process subscales. The scales are based on grounded theory generated from interviews and participant observations with nurses from 47 magnet hospitals. Items depict the steps or components of each process. Determination of the percentage of nurses in a unit who indicate that they can perform each step or component will allow identification of problem areas and what needs to be corrected to produce healthy, productive work environments.

In summary, nurses in all ICUs in this study, particularly those in NICUs, report healthy, productive work environments. Empirical study of ICUs and the possible linkages and relationships between nurse attributes, functional processes, and outcomes will advance theory and management practice. Study and analysis of subscale items will provide ideas and guidelines for assisting nurses and clinical units to achieve healthy work environments wherein the organization can be successful and nurses are happy and satisfied because they can give the best possible care to patients.

Acknowledgments

This multisite, evidence-based management practice initiative was a team effort. Nursing leaders and on-site investigators at each of the study sites contributed immeasurably to the support and coordination for this project. Study sites included The Miriam Hospital, Providence, Rhode Island; St Cloud Hospital, St Cloud, Minnesota; St Joseph’s Hospital of Atlanta, Georgia; University of Colorado Hospital, Denver; East Jefferson General Hospital, New Orleans, Louisiana; Providence-St Vincent’s Hospital, Portland, Oregon; Children’s Mercy Hospitals and Clinics, Kansas City, Missouri; and John C. Lincoln Hospital, Phoenix, Arizona. Our sincere gratitude and appreciation are expressed to the staff at those hospitals who participated in this study. Without their generous contribution of time, spirit, effort, and ideas this study could not have been done.

Financial Disclosures

This work was funded in part by a grant from the American Association of Critical-Care Nurses.

References

1. Shirey MR. Authentic leaders creating healthy work environments for nursing practice. Am J Crit Care. 2006;15(3):256-267.
2. American Association of Critical-Care Nurses. AACN standards for establishing and sustaining healthy work environments: a journey to excellence. Am J Crit Care. 2005;14(3):187-197.
3. Kramer M, Schmalenberg C. Development and evaluation of Essentials of Magnetism tool. J Nurs Adm. 2004;38(7/8):365-378.
4. Anthony MK. The relationship of authority to decision-making behavior: implications for redesign. Res Nurs Health. 1999;22(5):388-398.
5. Chaboyer W, Najman J, Dunn S. Factors influencing job valuation: a comparative study of critical care and non-critical care nurses. Int J Nurs Stud. 2001;38(2):153-161.
6. Cimiotti JP, Quinlan PM, Larson EL, Pastor DK, Lin SX, Stone PW. The magnet process and the perceived work environment of nurses. Nurs Res. 2005;54(6):384-390.
7. Kramer M, Schmalenberg CE. Staff nurses identify essentials of magnetism. In: McClure M, Hinshaw AS, eds. Magnet Hospitals Revisited: Attraction and Retention of Professional Nurses. Kansas City, MO: Academy of Nursing; 2002:25-59.
8. Alt-White AC, Charns M, Strayer R. Personal, organizational and managerial factors related to nurse-physician collaboration. Nurs Adm Q. 1983;8(1):8-18.
9. Prescott PA, Bowen SA. Physician-nurse relationships. Ann Intern Med. 1985;103(1):127-133.
10. Schmalenberg C, Kramer M, King C, et al. Excellence through evidence: securing collegial/collaborative nurse-physician relationships, part 1. J Nurs Adm. 2005;35(10):450-458.
11. Schmalenberg C, Kramer M, King C, et al. Excellence through evidence: securing collegial/collaborative nurse-physician relationships, part 2. J Nurs Adm. 2005;35(11):507-514.
12. Kramer M, Maguire P, Schmalenberg CE. Excellence through evidence: the what, when, and where of clinical autonomy. J Nurs Adm. 2006;36(10):1-12.
13. Boyle DK, Miller PA, Gajeski BJ, Hart SE, Dunton N. Unit type differences in RN workgroup job satisfaction. West J Nurs Res. 2006;28(6):622-646.
14. Boumans NPG, Landeweerd JA. Working in an intensive or non-intensive care unit: does it make any difference? Heart Lung. 1994;23(1):71-79.
15. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. An evaluation of outcome from intensive care in major medical centers. Ann Intern Med. 1986;104(3):410-418.
16. Wise LC. The erosion of nursing resources: employee withdrawal behaviors. Res Nurs Health. 1993;16(1):67-75.
17. Donabedian A. The Definition of Quality and Approaches to Its Assessment. Ann Arbor, MI: Health Administration Press; 1980. Explorations in Quality Assessment and Monitoring, volume 1.
18. Baggs JG, Schmitt MH, Mushlin AI, et al. Association between nurse-physician collaboration and patient outcomes in three intensive care units. Crit Care Med. 1999;27(9):1991-1998.
19. Ferrand E, Lemaire F, Regnier B, et al. Discrepancies between perceptions by physicians and nursing staff of intensive care unit end-of-life decisions. Am J Respir Crit Care Med. 2003;167(10):1310-1315.
20. Cummings GG, Hayduk L, Estabrooks CA. Is the Nursing Work Index measuring up? Moving beyond estimating reliability to testing validity. Nurs Res. 2006;55(2):82-93.
21. Choe J, Bakken S, Larson E, Du Y, Stone P. Perceived nursing work environment of critical care nurses. Nurs Res. 2004;53(6):370-378.
22. Aiken LH, Patrician PA. Measuring organizational traits of hospitals: the Revised Nursing Work Index. Nurs Res. 2000;49(3):146-153.
23. Kramer M, Hafner LP. Shared values: impact on staff nurse job satisfaction and perceived productivity. Nurs Res. 1989;38(3):172-177.
24. McClure M, Poulin M, Sovie M, Wandelt M. Magnet Hospitals: Attraction and Retention of Professional Nurses. Kansas City, MO: American Academy of Nurses; 1983.
25. Kramer M, Schmalenberg C, Hafner LP. What causes job satisfaction and productivity of quality nursing care? In: Moore T, Mundinger M, eds. Managing the Nursing Shortage: A Guide to Recruitment and Retention. Rockville, MD: Aspen; 1989:13-32.
26. Kramer M, Schmalenberg C. Learning from success: autonomy and empowerment. Nurs Manage. 1993:24(5):58-64.
27. Kramer M, Schmalenberg C. Magnet hospital nurses describe control over practice. West J Nurs Res. 2003;25(4):434-452.
28. Kramer M, Schmalenberg C, Maguire P, et al. Structures and practices enabling staff nurses to control nursing practice. West J Nurs Res. In press.
29. Kramer M, Schmalenberg C, Maguire P. Essentials of a magnetic work environment, part IV. Nursing2004. 2004; 34(9):44-48.
30. Goode CJ, Pinkerton S, McCausland MP, Southard P, Graham R, Krsek C. Documenting chief nursing officers’ preference for BSN-prepared nurses. J Nurs Adm. 2001;31(2):55-59.
31. Foley BJ, Jennings BM, Kee CC, Minick P, Harvey SS. Characteristics of nurses and hospital work environments that foster satisfaction and clinical expertise. J Nurs Adm. 2003;32(5):273-281.
32. Piazza IM, Donahue N, Dykes PC, Griffin MQ, Fitzpatrick JJ. Differences in perceptions of empowerment among nationally certified and noncertified nurses. J Nurs Adm. 2006; 36(5):277-283.
33. Ulrich BT, Lavandero R, Hart KA, Woods D, Leggett J, Taylor D. Critical care nurses’ work environment: a baseline status report. Crit Care Nurse. 2006;26(5):46-57.
34. Kramer M, Maguire P, Schmalenberg C, et al. Excellence through evidence: structures enabling clinical autonomy. J Nurs Adm. 2007;37(1):41-52.
35. Kramer M, Maguire P, Schmalenberg C. Nurses define and identify structures/practices that promote nurse manager support. Nurs Adm Q. In press.

Reprinted from American Journal of Critical Care, September 2007, pp 458-468 © 2007, AACN.

TOP