High turnover and low retention among nurses in hospitals are major problems in the United States,1 compounded by the demand for nurses and their increasing retirement rates.2 Research shows that various factors contribute to nursing vacancies and turnover, including unsupportive practice environments,3 long work hours, and excessive physical and psychological demands.4,5 In the early 1980s, as a response to problems with nursing retention and turnover, a task force was developed to identify organizational attributes of US hospitals that were successful in recruiting and retaining nurses despite a major national nursing shortage. A series of studies indicated that nurse turnover and vacancy rates in the Magnet hospitals were significantly lower compared with non-Magnet hospitals.6 This activity marked the beginning of the Magnet movement.7 The movement identified 14 attributes of successful hospitals, known as the forces of Magnetism. More recently, these were refined into 5 main components for Magnet accreditation: (1) transformational leadership, (2) structural empowerment, (3) exemplary professional practice, (4) new knowledge, innovation and improvement, and (5) empirical quality results.
The American Nurses Credentialing Center's Commission on Magnet Recognition(R) currently recognizes 348 healthcare organizations in the United States and 4 internationally. Magnet accreditation has evolved from a retention-focused activity into an avenue to promote a hospital's appeal to both consumers and the nursing workforce. Studies have identified attributes of Magnet hospitals that attract nurses: high autonomy, decentralized organizational structure, supportive management, and self-governance.7,8 Magnet hospital nurses rated decision making and control over practice and job satisfaction higher than nurses in non-Magnet hospitals.9-11 Magnet hospital nurses also reported a more supportive work environment, lower levels of burnout, and a higher intent to stay in their current job10,12 compared with nurses from non-Magnet hospitals. Brady-Schwartz9 also examined satisfaction with scheduling including hours and flexibility and found that Magnet facility nurses reported higher satisfaction, although comparison of hours worked and other schedule factors was not reported.
Despite these efforts, an ongoing concern for hospital nurses is their work schedule that includes long hours and extensive use of overtime.13 Although some nurses prefer 12-hour shifts, nurses were more fatigued when working 12 hours versus 8 hours.14 Excessive job demands are another major factor reported to adversely affect nursing retention. In a survey of more than 43,000 nurses in 5 countries, 17% to 39% reported that they planned to leave their job in the next year because of the demands.15 Nurses' work is physically and psychologically demanding,16,17 with few or no breaks taken during shifts. Excessive job demands can impair performance brought on by the resulting exhaustion or decreased cognitive acuity (vigilance, accuracy, and motivation),18,19 and they can increase the probability of musculoskeletal injury. Such injuries can then lead to turnover or loss of nursing staff. In a study, 20% of nurses changed jobs at least once because of an injury.20 Few studies indicate that nurses in Magnet hospitals experience a less demanding workload compared with their colleagues in non-Magnet hospitals.12
On the other hand, limited evidence indicates that Magnet hospitals have better patient safety and patient outcomes compared with their non-Magnet counterparts.10,21,22 For example, a survey of 3,016 nurses from 135 hospitals found that higher levels of autonomy, control, and collaboration were associated with higher perceived patient care quality.10 A nationwide study conducted of Magnet hospitals with 195 matched control hospitals found significant differences in mortality between Magnet and non-Magnet hospitals.21 Despite studies supporting the advent and continuation of Magnet accreditation as the standard for nursing excellence, controversy exists on whether Magnet status accomplishes the goal of improving the nursing work environment. McNeely23 emphasized that the typical nursing work environment contains high demands, low control, and low social support. Magnet hospitals focus on the organization rather than the individual nurse, suggesting that demands remain high among nurses providing frontline care at Magnet facilities.23 The purpose of our study was to compare working conditions including schedule, job demands, and practice environment between nurses working in Magnet and non-Magnet hospitals.
Methods
This is a cross-sectional secondary data analysis of 2004 data from the Nurses Worklife and Health Study.24 Institutional review board approval was obtained. The original study invited a probability sample of 5,000 actively licensed nurses from 2 US states, Illinois and North Carolina, to enroll. Of these nurses, 4,229 were sent surveys (138 had invalid addresses, and 633 declined), and 2,617 nurses (61.9%) responded, forming the baseline cohort of which 82.4% (n = 2,156) participated in the third wave of data collection, which was used in this analysis. Demographic and job characteristics of the participants were comparable to distributions of national samples of US nurses in the national sample survey of nurses.25 The sample was 95% female with a mean age of 45 years.
For this analysis, we restricted the sample to those working in an acute-care hospital in 2004, excluding retirees (n = 210), nonhospital nurses (n = 862), and nurses who did not specify the name of their hospital (n = 233). This yielded a final sample of 837 nurses working in 171 acute-care nonfederal, general medical-surgical hospitals. The hospitals were designated as Magnet (n = 14) or not (n = 157) based on 2005 American Nurses Credentialing Center accreditation status. The Magnet hospital status as of 2005 was used to divide the nurses into groups, after comparing these findings to those obtained using the 2004 Magnet hospital designation. As it turns out, results were quite similar, but with the 2005 Magnet hospital status, the groups of nurses were more balanced in terms of size, which is preferable from a statistical standpoint. In addition, hospitals achieving Magnet Recognition in 2005 were of necessity, already on the journey toward Magnet status during 2004, so that for nurses working in such hospitals, the experience would likely strongly resemble a Magnet-designated facility. We therefore present results using the 2005 Magnet designations. Nurses working in these hospitals were divided into Magnet (number of nurses = 162) and non-Magnet groups (number of nurses = 675) and were compared on the following characteristics: work schedule, job demands, and practice environment.
We also created a 3-level variable comparing nurses working in Magnet hospitals designated in 2004, to nurses working in hospitals designated in 2005 Magnet hospital nurses, versus nurses working in non-Magnet hospitals (as of 2004-2005), similar to that of Ulrich et al.26 These analyses showed minimal differences from the other analyses; therefore, we included results from only the 2005 Magnet hospital designation for the reasons indicated above.
Study Variables
Demographics and descriptive characteristics were assessed including education and type of unit worked, along with measures of working conditions, with measures, including reliability and validity information, summarized in Table 1. For work schedule, nurses reported on their typical schedule for the past 6 months on average, to minimize responses covering an atypical work period.27 Nurses were asked to report the hours they actually worked, as opposed to those they were scheduled to work. Variables measured covering work schedule included those from the Work Schedule Index: hours per day, hours per week, days per week, weekends per month, breaks of 10+ minutes, full- versus part-time, and shift (days only vs other). Additional schedule items included frequency of working 13+ hours, of having less than 10 hours off between shifts, of working on days off/vacation days or while sick, and whether their jobs included mandatory overtime and required on-call (yes/no).
Job demands were measured by the Job Content Questionnaire, which has been shown to validly measure the psychosocial work environment of the organization based on individual responses28 in multiple occupations and nurse samples.32,33 Psychological demands included working very hard, working very fast, excessive amounts of work, long periods of intense concentration, having enough time to get the job done, tasks that are interrupted before completion, and waiting on work from others.28 Physical demand items reflected the duration, intensity, and frequency of exposure based on 12 items that measured, for example, heavy lifting and awkward postures.34
Nurse practice environment measures included autonomy, support, perceived patient safety culture, and job satisfaction, with items from the Nursing Work Index-Revised (NWI-R),29 the Job Content Questionnaire support domain,28 the Patient Safety Center of Inquiry Culture Survey,30 and the Hospital Survey on Patient Safety Culture.31 The NWI-R items measured physician-nurse relationships, if nurses have to do things against their judgment and if they have adequate support systems, along with nurse participation in cost controls and nurse input into equipment selection. Job support items included whether "my supervisor considers my viewpoint, can suppress personal biases, treats me with kindness, and deals with me truthfully." Other support items were "coworkers can be relied upon when I need help" and "coworkers are helpful in getting the job done." Safety items were "our procedures and systems are good at preventing errors, it is just chance that more serious mistakes don't happen in my workplace, staff freely speak up if something can negatively affect patient safety, management seems to care more about cutting costs than improving safety, patient safety is sacrificed to get work done" and "safety decisions are made at the proper level by those most qualified." Overall job satisfaction was assessed using 1 Likert-type item, as such measures correlate highly with multiple-item measures.35
Data Analysis
Analysis was performed using SPSS version 15.0 (SPSS Inc, Chicago, Illinois). Univariate, descriptive statistical analyses were conducted, and the mean and SDs of the key variables were calculated. To compare characteristics of nurses working in Magnet hospitals to those working in non-Magnet hospitals, we used t tests for continuous variables (ie, age) and Pearson [chi]2 for categorical variables (eg, sex) with Fisher exact tests for cell sizes less than 5. Bivariate analyses were then conducted using the ttest in a general linear model for continuous variables. To accommodate clustering of nurses in a hospital, the Huber-White sandwich estimator of variance was used to obtain robust SEs and variance estimates.
Results
Nurses who worked in Magnet and non-Magnet hospitals did not differ in terms of demographic characteristics including age, sex, marital status, educational level, and unit type, whereas for race/ethnicity the proportion of nurses of color working in Magnet hospitals was significantly lower (8.6%) than among nurses working in non-Magnet hospitals (16.1%) ([chi]2 = 5.964, P =.018) (Table 2). Overall, there were very few differences in terms of their working conditions (Tables 3,4). Those working in Magnet hospitals were less likely to report that their jobs contained mandatory overtime and on-call than those in non-Magnet hospitals, although there were no differences in hours worked per day or per week. For job demands, there were no differences in psychological demands reported by nurses working in Magnet versus non-Magnet hospitals. Physical demands were lower among nurses in Magnet hospitals compared with those working in non-Magnet hospitals, with the mean for Magnet hospital nurses equal to 30.1 versus 31.0 for non-Magnet hospital nurses (t = 2.140, P =.034). On all other measures of nursing working conditions including nursing practice environment (NWI-R), patient safety culture, and overall job satisfaction, there were no significant differences between nurses working in Magnet or non-Magnet facilities.
Discussion
In our study comparing working conditions reported by nurses working in Magnet and non-Magnet acute-care hospitals, we found very few differences. We did find that nurses who worked in Magnet hospitals were less likely to report having mandatory overtime and on-call as part of their jobs, although reported hours worked did not differ. Nurses working in Magnet hospitals also reported lower physical demands. However, this statistically significant difference is likely of little practical significance, as physical demand means were within 1 point on a scale ranging from 12 to 48. A finding of few differences may not be totally unexpected. As noted previously, the Magnet journey focuses on structures and processes that do not typically address the work schedules and job demands of nurses. Our findings are similar to those of Ulrich et al,26 who found few differences overall among nurses working in Magnet hospitals, although they did note that nurses in Magnet-aspiring, Magnet-designated, and non-Magnet hospitals differed on some job-related conditions.
Our results suggest that working in a Magnet-designated facility does not necessarily mean that nurses perceive better working conditions, although working conditions have been found to be major factors in nurse retention.1,36 Alternatively, positive practice environments that include adequate staffing, organizational support, and satisfaction with supervisor have been identified as key elements in staff retention, and nurses working in both Magnet and non-Magnet hospitals have been beneficiaries of these changes.4,37 As noted by Hassmiller and Cozine,38 hospitals must be more appealing and functional if they want to recruit and retain nurses, an implicit assumption of the Magnet hospital movement. In addition, nurses who have left hospitals because of adverse schedules and other hazardous job conditions prefer working in facilities with predictable work hours and schedules.39 If these factors are addressed, perhaps hospital workforces can stabilize by attracting and retaining nurses in all types of hospitals.40
Results should be considered in light of the study limitations. All data were self-reported, thus subject to errors regarding recall and biases due to socially desirable responding. Nonetheless, as the context for this analysis was not known to participants, it is unlikely that their responses contained bias related to the Magnet status of their workplace. Misclassification of facilities is possible because Magnet Recognition takes time to achieve, and nurses in hospitals seeking this designation may experience changes prior to Magnet Recognition. Our additional analyses of the data with 2004 designations and with the data separately by state (Illinois and North Carolina) found only minor differences, suggesting this does not explain the limited differences between the groups. Issues with level of analyses are also acknowledged. Despite our best efforts, assessment of working conditions can reflect the nurses' experiences in their unit rather than reflecting the overall organization. These were secondary data so the information was limited to what had already been collected. We did not include control variables as these generally attenuate differences between groups and reduce power to detect differences that are there.
Studies examining the impact of Magnet and other nurse retention-oriented workplace solutions are important, to provide evidence-based information for policy makers and administrators. There are other reasons why nurses remain in or leave jobs besides Magnet characteristics; many reasons are related to working conditions such as we have examined in this study. Although nursing shortages have diminished in certain areas because of economic downturns, employers, nursing organizations, and labor representatives should focus attention on the important issue of working conditions and designate efforts toward developing long-term solutions. Furthermore, research should focus on how to effectively improve nurses' conditions and in turn improve their quality of life.
References