Kangas, Sandra MS, RN; Kee, Carolyn C. PhD, RN; McKee-Waddle, Rebecca MSN, RN
An increase in for-profit acute care hospitals, a competitive healthcare market, the state of the national economy, and increasingly sophisticated consumers drive the goal of providing cost-efficient and effective healthcare. Attracting patients and managed-care corporations has become a priority of hospitals so that a patient census large enough to produce a profit margin can be maintained. Almost without exception, hospitals are conducting patient satisfaction surveys, creating patient advocacy departments, and, in some cases, offering employee incentives based on patient satisfaction statistics (T. Riley, personal communication, May 15, 1998).
Because the personnel budget is the largest expense in the overall hospital budget and nursing personnel represent the largest group in the personnel budget, containing costs frequently means reducing nursing personnel. This produces a strain on goals to maintain quality patient care, thus threatening patient satisfaction and the hospital's ability to attract patients. Recognizing that personnel turnover is expensive both monetarily and in terms of patient satisfaction, nursing administrators also have focused on retaining nursing personnel as a way to contain costs. Nurses' job satisfaction is viewed as integral to nurse retention.
In the 1970s and 1980s, hospitals responded to the findings of job satisfaction studies and began to experiment with organizational structures and management processes in efforts to improve job satisfaction. Terms such as participative management, shared governance, committee structure, primary nursing, case management, and staff involvement became commonplace. Changing the organizational structure as a way to improve the work environment or organizational culture was seen as a way to improve nurse's job satisfaction and enhance retention.
The purpose of this study was to identify differences and relationships among the job satisfaction of registered nurses (RNs), patient satisfaction with nursing care, nursing care delivery models, organizational structure, organizational culture, and general background and demographic information.
A number of studies examining nurses job satisfaction, patient satisfaction with nursing care, nursing care delivery models, and organizational structure and culture are reported. The studies included in the literature review demonstrate the complexity of these issues and their importance to quality, cost-efficient patient care provided in an environment that supports professional practice.
Nurses' Job Satisfaction
The issue of staff nurse retention is an enduring problem in nursing and was the initial stimulus generating interest into the work satisfaction experienced by nurses.1 Using meta-analysis, Blegen1 examined variables affecting nurses' job satisfaction. Her study included data from 48 studies with more than 15,000 participants. The results indicated that job satisfaction for nurses was correlated negatively with stress and was correlated positively with commitment to the organization. Communication with supervisors and peers, autonomy, recognition, routinization, fairness, and locus of control were correlated less strongly with job satisfaction.
Recent studies support the importance of organizational commitment and structure to the job satisfaction of nurses2-4 as well as to the job satisfaction of nurse managers.5 Blegan and colleagues3 found that staff nurses and head nurses agreed that the most desirable organizational structure was one that supported autonomous decision making by staff nurses. Acorn and colleagues5 reported that for nurse managers, decentralization had a positive effect on perceived autonomy, job satisfaction, and organizational commitment.
Patient Satisfaction with Nursing Care
Because quality patient care is considered essential to institutional survival, it is an attribute sought after by hospitals and emphasized by administrators.6-10 A study comparing patient perception of quality care and patient satisfaction with quality care found that failing to meet expectations of quality was synonymous with poor quality and resulted in patient dissatisfaction.11 Patient satisfaction with nursing care was found to be an important predictor of overall satisfaction with hospital care.12
A recent study on quality improvement compared nurses' and patients' definitions of quality and the values each placed on different aspects of care.10 The values held by patients were found to be different from the values that health professionals thought were held by patients. Young and colleagues10 suggested that the organizational culture of the unit, driven by the values of the nurse manager and unit staff, may have reinforced differences between patient and care-giver values. As a result of this study, there was an effort made to transform the unit culture to support aspects of care valued by the patients.
Nurses' Job Satisfaction and Patient Satisfaction
In 1992, Henry13 found that RN turnover was inversely related to patient satisfaction. She suggested that administrative actions be examined and evaluated for their effects on patient and nurse-related outcome measures such as satisfaction with nursing care and staff turnover.
Worthington14 recounted an interview with John Durant, MD, Vice President for Health Affairs at the University of Alabama at Birmingham Hospital. Durant described findings from a hospital study where positive staff attitudes were the most consistent predictor of patient satisfaction. One nurse manager in this study commented that patient satisfaction decreased when staff morale decreased.
Grindel and colleagues15 found that quality patient care occurred in practice environments with high degrees of patient satisfaction, physician satisfaction with patient care, and nurse (job) satisfaction. Using a combination of assessment tools, patient, physician, and nurse satisfaction were analyzed together to develop strategic initiatives for improvement of the overall practice environment.
Nursing Care Delivery Models
In Thomas' study2 describing how nurses and non-nurses perceived their work in primary, team, and functional practice settings, little difference was noted in the ways in which nurses and non-nurses perceived their working environments. Where primary nursing was practiced, both nurses and non-nurses perceived greater supervisor support, autonomy, physical comfort, and less work pressure when compared with their counterparts working in places at which team and functional nursing were practiced. Kramer and colleagues16 believe that the motivational effects of primary nursing improved the competence of individual nurses.
Case management also has been presented as a solution for empowering nurses and giving authority, control, and creative expression to their professional practice.17,18 Weinstein19 commented that institutions using this model have reported cost savings as well as increased patient, nurse, and physician satisfaction and improved quality of care. Ethridge and Lamb20 noted that care coordinated by a case manager resulted in shorter hospital stays, and thus, decreased costs. One case management model was cited as achieving a 40% reduction in patient care costs.18
In his capacity as a nurse administrator, Porter-O'-Grady21 sought to create organizational structures that ameliorated high turnover and nursing staff dissatisfaction. He was instrumental in designing and popularizing the shared governance philosophy. Shared governance uses a decentralized approach to management in which staff nurses make decisions on issues impacting their work and working environment, professional development, and personal fulfillment.
Gustin22 compared the job satisfaction of nurses working in a shared governance environment to that of nurses working in a traditional structure and found that job satisfaction was significantly higher in the shared governance hospital. In contrast, a later quasiexperimental study examining the effects of shared governance on nurse and non-nurse perceptions of work and work environment found that initiating shared governance did not significantly influence job satisfaction, anticipated turnover, and perceived effectiveness.4 Further, the sense of increased autonomy associated with greater influence in decision making was not sustained over time.
A corporate culture of excellence is a major determining factor in nurse job satisfaction, productivity, job attraction, and retention.6,9 As described in the Magnet Hospitals study,23 a culture of excellence includes support for education, self-governance, and opportunities for specialized practice. In cultures of excellence, environments are created that eliminate internal nursing shortages by successfully attracting and retaining satisfied, productive nurses.16
Studies Linking Job Satisfaction, Organizational Structure, and Organizational Culture
The American Association of Critical Care Nurses initiated a study to explore the association between costs and the effectiveness of nursing care delivery in critical care units.24 The authors noted that the literature clearly documented connections between hospital structures with positive organizational climates and outcomes such as high job satisfaction and morale and low turnover. These positive organizational outcomes occurred in climates characterized by staff "... participation in decision-making, clear policies and rules, autonomy in carrying out policies, and a variety of coordinating mechanisms."24(p221) Findings from their study linked low mortality, no new complications, and patient satisfaction to highly rated nursing performance, high nurse-physician collaboration, positive organizational climate, job satisfaction, and high morale.
As evident in this literature review, practice models, organizational structure, and organizational culture have been studied in different ways. When the outcome variables of nurses' job satisfaction and patient satisfaction with nursing care are considered, complexity increases even further. However, it is important to discover the ways in which these factors affect each other. Institutions have not been loath to try new systems of care, but they do so without a substantial body of literature to support a particular course of action. The present study continued and extended research in this area.
1. Are there differences in nurses job satisfaction and organizational culture in hospitals with different nursing care delivery models or different organizational structures?
2. Is the job satisfaction of nurses affected by nursing care delivery model, organizational culture, and other background and demographic factors?
3. Are there differences in patient satisfaction with nursing care in hospitals with different nursing care delivery models or different organizational structures?
4. Is patient satisfaction with nursing care affected by nursing care delivery model, previous hospitalizations, length of stay on a unit, and other patient demographic factors?
A correlational descriptive survey design was selected for this study.
Three hospitals representing three different nursing care delivery models, including team nursing, case management, and primary nursing, were selected. In the team nursing model, RNs, licensed practical nurses (LPNs), and nursing assistants formed a team to provide care for a group of patients. Nursing tasks are divided among team members according to skill level needed and qualifications of the person giving care. In the case management model, staff nurses were assigned specific patients to follow and monitor throughout their hospital stay. Parameters for nursing care were guided by clinical pathways in the hospital where the case management model was employed. Where primary nursing was practiced, RNs cared for consistent groups of patients over time as acuity and nurse scheduling allowed. Nursing assistants also provided care but were few in number and assigned according to where help was needed the most on a given day.
The organizational structure of two of the three hospitals was traditional, with a separate department for nursing services, a director of nursing, nurse managers and charge nurses for the units, and a hierarchical structure of authority. The third hospital used a shared governance model, with a vice president for nursing and staff participation in decision making. The shared governance organizational structure used council groups composed of mostly staff representatives for discussion and decision about policies and procedures. Nonstaff council members (administrators and clinical nurse specialists) acted in an advisory capacity only. The core philosophy in shared governance is that professional nurses should work collaboratively together to provide nursing care and to determine all related activities that support nursing practice, such as the development of nursing standards, quality assurance, research, and management.
The resulting nursing care delivery model and organizational structure combinations were the following: 1) Hospital 1 had a traditional, hierarchical organizational structure and practiced team nursing; 2) Hospital 2 also had a traditional, hierarchical organizational structure but used the case management model of nursing care delivery; and 3) Hospital 3 had a shared governance organizational structure and primary nursing as the nursing care delivery model.
Nurses and patients were selected from each of these institutions. Inclusion criteria for the RNs were that they had a minimum of 6 months' nursing experience, had worked in their current hospital for at least 6 months, and were on adult medical-surgical units, inpatient critical care areas, or critical care step-down units. Inclusion criteria for the patients were that they were on the same units as the selected RNs and were physically and mentally able to answer the questionnaire.
A contact person was identified at each hospital to assist with identifying appropriate units and RNs. Systematic random sampling was used to select the RN sample from lists of all RNs assigned to a unit, regardless of shift worked. Next, nurse managers on the units identified patients meeting the criteria from the same units on which these RNs worked so that there would be approximately equal numbers of patients and nurses. For example, if four registered nurses were selected from unit A, then four patients from unit A also would be selected. Power analysis with an effect size of 0.2, an alpha of 0.05, and a level of 90% indicated that the required sample size was 102 RNs and 102 patients.
Instrument for the Nurse Sample
The instrument for the RNs consisted of a demographic and background data form, a job satisfaction scale, and a measure of organizational culture. Total time to complete the instrument was 15 to 20 minutes.
Demographic Data Form
Background information requested from the nurses included age, gender, education, race, number of years in nursing, part- or full-time employment, whether on a medical-surgical or intensive care unit, length of time at the present hospital, and frequency of being in charge.
Nurses Job Satisfaction
Torres25 developed a measure of nurses' job satisfaction from work previously done by Bullough26 and Everly and Falcione.27 The Nurse Job Satisfaction Scale is a 26-item instrument that uses a 5-point Likert scale ranging from 1 to 5 points. Items reflect areas deemed important to nurses job satisfaction and include descriptors that reflect autonomy, ability to be creative, work load, morale, and opportunity for advancement. Adjectives such as high/low (for autonomy and ability to be creative), heavy/light (work load), and good/poor (morale and opportunity for advancement) are used for respondents to answer items reflecting various dimensions of work experiences. Total score range is from 26 to 130; the lower the score, the greater the degree of job satisfaction. Torres25 reported face validity, a content validity index of 0.92, and Chronbach's alpha reliability as 0.83.
Wallach28 refined the work of both Litwin and Stringer29 and Margerison30 to develop an Organizational Culture Index. This index categorizes organizational culture into three dimensions: bureaucratic, innovative, and supportive cultures.
Bureaucratic cultures28 are hierarchical with structured lines of responsibility and authority. Control and power are clearly evident. A high score on bureaucracy means the organization is power oriented, cautious, established, solid, regulated, ordered, structured, procedural, and hierarchical.28(p32) Innovative cultures28 are characterized by excitement and dynamism with challenging, risky, and creative environments where entrepreneurs do well. In innovative environments, burnout and stress are common occupational hazards. Supportive cultures28 have pleasant and harmonious working environments. Interactions are friendly, and people help one another. Supportive cultures are open, with the environments resembling extended families. They are trusting, safe, equitable, sociable, encouraging, open, relationship oriented, and collaborative.
The Organizational Culture Index has 24 items divided into three subscales, one for each dimension. Each subscale has eight items that are answered on a 4-point Likert scale ranging from 0, "does not describe my unit," to 3, "describes my unit most of the time." The score of the Organizational Culture Index is the sum of the responses for the items in each subscale, with a possible subscale range of 0 to 24. Total scale scores are not computed because all work cultures have some elements of each dimension. The dimension with the highest score is considered the dominant one for that particular environment. Koberg and Cushmir31 reported subscale alpha coefficients from 0.75 to 0.91.
Instrument for the Patient Sample
The instrument for the patients consisted of a demographic data form and a measure of satisfaction with nursing care. The instrument required a fifth-grade reading level and took 10 to 15 minutes to complete.
Demographic Data Form
Data requested from patients included age, gender, education, length of time on the unit, and previous admissions over the past year.
Patient Satisfaction with Nursing Care
In 1975, Risser32 developed a scale to measure patient satisfaction with nurses and the nursing care given in primary care settings. The scale has 25 items and includes three dimensions of care: 1) the technical-professional area, which includes the knowledge and expertise of the nurse when providing care (7 items); 2) the educational relationship area, which refers to information-giving behaviors of the nurse (7 items); and 3) the trusting relationship area, which assesses the nurses interest in the patient's feelings and problems (11 items). Alpha coefficients in the initial study ranged from 0.637 for the technical-professional subscale to 0.912 for the total scale.
In 1982, Hinshaw and Atwood33 slightly revised one item so that the scale would be appropriate for use in inpatient settings. In their studies using the revised instrument, they found alpha coefficients of 0.80 to 0.98 for all subscales except for one study in which the education subscale was 0.44. Construct validity was established' using a multitrait-multimethod approach. Convergence scores were 0.80 to 0.90, and discriminance scores were between 0.06 and 0.63. Torres25 also used this instrument and reported reliability coefficients ranging from 0.90 to 0.93 and a content validity index of 0.92.
A 5-point Likert scale with descriptors of "strongly agree" to "strongly disagree" is used to respond to the questions. Some items are reversed to avoid response set bias. Score range from 25 to 125 for the entire scale. The lower the score, the greater the satisfaction with nursing care.
Data Collection Procedures
Institutional Review Board approval was obtained from the authors' university and from each of the three hospitals. The study was classified as exempt, and all those participating were assured that confidentiality and anonymity would be maintained.
Once approval to conduct the study was received, RNs and then patients were selected according to the procedure described previously. Each randomly selected RN was contacted personally by one of the investigators, and she or he received a packet containing an introductory letter describing the study and parameters of informed consent, a form that requesting demographic information, and the questionnaires on organizational culture and job satisfaction. Return of the completed questionnaire served as the RNs consent to participate. Once patients were identified by the nurse manager, they were approached regarding willingness to participate in the study. Patients who agreed also received a packet containing an introductory letter describing the study and parameters of informed consent, a demographic data form, and a questionnaire about their satisfaction with nursing care. Completion of the questionnaire by patients also served as their consent to participate.
A total of 92 RNs completed the survey. As shown in Table 1, most of the nurse respondents were women, in their 30s, white, and had been in nursing for more than 5 years. The majority had been on the same unit for more than 5 years, worked in medical-surgical units, were full-time, and at least occasionally assumed charge nurse duties. Most had changed jobs at least two times since licensure, and most had either associate and baccalaureate degrees in nursing.
A total of 90 patients completed the survey. As shown in Table 2, the patient population was about evenly divided by gender, although there were slightly more men than women. Most were older than 44 years of age and educated at the high school level or higher. Length of stay was 4 days or less for most respondents, although approximately 34 (38%) had been hospitalized for more than 4 days. Thirty-four respondents (38%) had been hospitalized at least once during the 12 months before the current admission.
Scale scores for nurses' job satisfaction, organizational culture, and patient satisfaction with nursing care for all hospitals combined are shown in Table 3. The possible range and midpoint for each scale are shown as well.
Nurses Job Satisfaction Scores
Low scores on the job satisfaction scale indicate higher levels of job satisfaction. Scores on job satisfaction for all nurse respondents together ranged from 40 to 105 with a mean of 69 (SD = 12). Mean job satisfaction scores for the hospital with team nursing was 70 (SD = 15); for the hospital with case management, it was 68 (SD = 11); and for the hospital with primary nursing, it also was 68 (SD = 11). Reliability for this study using Chronbach's alpha was 0.85.
The higher the score on the organizational culture subscale, the more that characteristic is present in the environment. For nurse respondents in this study, all hospitals had about equal degrees of bureaucratic, innovative, and supportive subcultures present. Chronbach's alpha for the bureaucratic subscale was 0.73; for the innovative subscale, the coefficient was 0.78; and for the supportive subscale, Chronbach's alpha was 0.90.
Patient Satisfaction with Nursing Care Scores
Also as shown in Table 3, the total score mean and most subscale score means were at or below midpoint, indicating that patients were satisfied with their nursing care overall. Reliability for the total scale was 0.95; Chronbach's alpha for the technical/professional subscale was 0.83; for the trusting subscale, the coefficient was 0.89; and for the education subscale, the coefficient was 0.84.
Research Question 1: Differences in Nurses Job Satisfaction with Different Care Delivery Models and in Different Organizational Structures
Analysis of variance was used to test for differences in the job satisfaction of nurses practicing with different nursing care delivery models (team nursing, primary nursing, and case management) and then by the two different organizational structures (traditional and shared governance). No significant differences were found for either analysis although analysis of variance for job satisfaction differences by care delivery model approached significance (P = 0.077). Mean job satisfaction score of nurses working in primary care was 50; for those in case management, the mean score was 56; and for those in team nursing, the mean score was 58.
Research Question 2: Factors Affecting Nurses Job Satisfaction
Multiple regression was used to assess factors that might predict nurses job satisfaction. Independent variables were selected for inclusion in the model based on their bivariate correlation with the dependent variable. The variables included were organizational structure, nursing care delivery model, bureaucratic subscale scores, innovative subscale scores, supportive subscale scores, frequency of being in charge, and type of unit worked (either general medical-surgical unit or critical care unit). As displayed in Table 4, perceiving the environment as supportive and working in a critical care unit were significant predictors of nurses' job satisfaction scores. These two variables together predicted 55% of the variance in job satisfaction. The standardized beta for the supportive culture subscale was roughly twice that of type of unit.
Research Question 3: Differences in Patient Satisfaction with Different Care Delivery Models and in Different Organizational Structures
Analysis of variance also was used to test for differences in patient satisfaction by nursing care delivery model and by organizational structure. No significant differences were found for either analysis. Scores of patients receiving care in the primary delivery model were slightly lower, indicating that they were somewhat more satisfied, but the difference was not significant.
Research Question 4: Factors Affecting Patient Satisfaction
Independent variables selected for inclusion in the multiple regression equation were nursing care delivery model; organizational structure; patient age, gender, and educational level; and length of time they had been a patient on the unit. Only nursing care delivery model and length of time on the unit (inverse correlation) were significant in predicting patient satisfaction with nursing care (P = 0.008). Patients receiving nursing care in the primary care delivery model expressed more satisfaction. The longer patients had been on the unit, the less satisfied they were. The adjusted R2 was 0.097, so that only 10% of the variance in patient satisfaction was accounted for by delivery model.
Discussion and Implications
Overall, scores on the Nurses Job Satisfaction scale indicated that nurses in this study were slightly more satisfied with their jobs than not. Distance from the midpoint was not impressive, however, and there were no differences in job satisfaction scores by nursing care delivery model. This was an unexpected finding because the literature supported both primary nursing and case management as important contributors to job satisfaction. Although care delivery models theoretically are distinct, perhaps the realities of clinical practice mute any real differences and thus negate differential effects on job satisfaction. However, care delivery models simply may not be an important influence on nurses job satisfaction at all. The process of implementing a structural variable such as nursing care delivery model may be more important than the model itself. In other words, if RNs perceive themselves to be valued employees who are critically important to the institution and to their patients' well-being, then structural changes in the work environment are handled with equanimity. Conversely, if the opposite is true, and RNs feel devalued with little of importance to contribute to patient care, then differences in nursing care delivery models would have no effect on job satisfaction.
Findings from the multiple regression analysis lend some support to the notion that elements other than nursing care delivery model affect job satisfaction. In fact, the culture (supportive) in which care was delivered subsumed the type of unit (critical care or medical-surgical) in degree of impact on nurses' job satisfaction. This, too, was a surprising finding.
In this study, unit type was either a general medical-surgical unit or an intensive care unit. Working in intensive care units allows nurses to specialize in highly sophisticated areas of practice that demand expertise in narrowly defined fields. Studies have shown that specialization is an important and even major contributor to nurses' job satisfaction. 34 Specialization allows in-depth development of knowledge and assessment, technical, and intervention skills in highly technical areas such as intensive care units.
However, working in medical-surgical units does not promote specialization because a broad expanse of knowledge and skills is required for these units. Medical-surgical units require expertise in organizational skills and general, rather than specialized, knowledge because care is delivered to more patients with greater variability in acuity level, disease category, and functional status. Organizational and administrative skills tend to be invisible and unrecognized skills, evident only when tasks are not done or incident reports increase in severity and frequency.
Specialization as measured by the variable type of unit in this study was important to nurses' job satisfaction, but a supportive culture was roughly twice as important. Nursing is a service profession, and skill ultimately is dependent on the ability to interact with others to achieve patient goals. Elements present in supportive work cultures may be those essential to promoting the ability of nurses to practice at high levels. Supportive environments may enhance self-recognition of successful job performance and give a sense of doing well, thus leading to higher levels of job satisfaction. McNeese-Smith35 found that job satisfaction was influenced strongly by recognition, praise, and support of the team. These characteristics are inherent in Wallach's definition of supportive cultures. A supportive organizational culture may contribute to the development and enhancement of a professional sense of self-esteem, and a professional sense of self-esteem is linked logically to job satisfaction.
Overall, patients were satisfied with the nursing care they received. Approximately half of the patients were hospitalized for 3 days or less when they answered the survey, and all were hospitalized at the time the survey was given. This may have affected responses one way or the other. Patients may have reported more satisfaction because they still were hospitalized and felt vulnerable. Alternatively, they may have reported less satisfaction because of illness and the accompanying unpleasant symptoms.
Two of the three patient satisfaction subscale scores were at or below midpoint indicating average or slightly better satisfaction. The subscale score for education was slightly above midpoint indicating some dissatisfaction in this area. Although patients indicated that they were not especially satisfied with the education received from nursing staff, little teaching actually may have occurred or patients may have known that they were to be discharged and were fearful. Patient education may not have been done yet, or discharge anxiety may have been so high that instructions simply were not absorbed. In either case, however, patient education is a critical area in terms of complete recovery and health maintenance. Ways of effectively conveying health information to patients in this era of short hospitalization times need to be identified and incorporated into hospital environments.
The total multiple regression variance accounted for in patient satisfaction with nursing care scores was small. Both nursing care delivery model and time on the unit were equally, although minimally, important. Although the satisfaction scale was intended specifically to address patient satisfaction with nursing care, patients may have answered from the perspective of how satisfied they were with the entire hospitalization experience and with any personnel, nurses and non-nurses, encountered up to the time that they answered the questionnaire. If this were true, then variables identified for analysis did not captured the full range of factors important to the whole hospital experience. Also, there may be less recognition now of who the nurses on the unit are so that they, and the importance of the nursing care that they are responsible for, seem invisible. The primary care delivery model might make nursing care more visible to patients than do other models even though quality may not be at all different. Regardless, the variance in patient satisfaction scores was largely unexplained.
Although random sampling was used to select the RNs in the study, purposive sampling was used to select the three hospitals and the patients so that bias is possible. Nurses were not matched with the patients for whom they cared, so analysis on the relationship between nurse satisfaction and patient satisfaction could not be done at the individual level. Patients were given the satisfaction with nursing care questionnaire while still hospitalized, which may have affected responses in unknown ways. As a whole, the measures used were sufficiently valid and reliable.
The cultural milieu of hospitals has received limited examination as a variable important to nurses' job satisfaction. The Magnet Hospital study23 demonstrated how important certain aspects of culture were to job satisfaction, but the perspective in that study was somewhat different. The organizational culture measure used in the present study was more subjective and captured broad characteristics of the working climate that affect multiple aspects of work performance. The significance of a supportive environment to the nurses in this study indicates the need for more extensive research into this aspect of job satisfaction.
The nature of the link between nurses' job satisfaction and patient satisfaction with nursing care is elusive and continues to need explication. Assuming that nurses who are satisfied with their jobs provide good patient care is not sufficient. The ways in which supportive environments affect specific nurse-patient interactions need to be pinpointed. The hospital structural variables thought to influence patient satisfaction in this study did not. If the pleasant, encouraging, extended family-like atmosphere described in Wallach's supportive organizational cultures is important to nurses, then it may be equally important to patients. It may be that in units in which personnel view each other as members of an extended family, the "family circle" expands to include patients. Devising ways to examine this notion might prove fruitful. Nurses and the unit subculture that they create and maintain may be pervasive and critical factors in nurses' job satisfaction and patient satisfaction with nursing care.
Nurses' job satisfaction and patient satisfaction with nursing care are important issues for nurse executives and nurse managers. Nursing administrators want a nursing staff that is productive, dynamic, innovative, and pleased with the work that they do. Nursing administrators want patients to feel that nurses care about their well-being and want them to get better. Nurses' job satisfaction is the key to creating work environments that meet these goals. Attending to process variables such as promoting supportive environments is one likely way to achieve them.
The authors thank T. Riley, RN, and Patricia Hall, RN, PhD, for their help in facilitating this project and with data collection.
© 1999 Lippincott Williams & Wilkins, Inc.