Abstract
Aims
To test the associations of safety practices as reported by nurses and their respective unit supervisors with job satisfaction.
Background
Psychosocial workplace factors are associated with job satisfaction; however, it is unknown whether nurses and supervisors accounts of safety practices are differentially linked to this outcome.
Methods
Cross-sectional study design including nurses (n = 1052) nested in 94 units in two hospitals in Boston (MA, USA). Safety practices refer to the identification and control of occupational hazards at the unit. Safety practices were measured aggregating nurses’ responses per unit, and supervisory levels. Individual’s job satisfaction for each nurse was the response variable.
Results
Supervisors assessed safety practices more favourably than their unit nursing staff. Adjusted random intercept logistic regressions showed that the odds of higher job satisfaction were higher for nurses at units with better safety practices (OR: 1.67, 95% CI: 1.04, 2.68) compared with nurses at units that averaged lower safety practices. Supervisors’ reports of safety practices were not correlated with the job satisfaction of their staff.
Conclusions
Adequate safety practices might be a relevant managerial role that enhances job satisfaction among nurses.
Implications for nursing management
Nursing supervisors should calibrate their safety assessments with their nursing staff to improve nurses’ job satisfaction.
Keywords: job satisfaction, multilevel studies, nurse management, safety practices
Introduction
Job satisfaction is a critical outcome for managing a health care workforce. Job satisfaction is related to lower turnover rates and increased hospital quality (Lu et al. 2005, Aiken et al. 2012). Research has linked job satisfaction with salient psychosocial characteristics at health care organisations such as autonomy and control (Elovainio et al. 2000, Aiken et al. 2002), leadership styles (Utriainen & Kyngas 2009) and an ethical and caring organisational culture (Gershon et al. 2007). However, the impact of safety-related factors on job satisfaction among nurses has been less studied, despite the fact that nursing is a high-risk occupation, exposed to multiple hazards that may result in work-related diseases and injuries. This inquiry is significant because evidence from other workforces indicates that organisational policies and practices related to safety are associated with fewer occupational injuries and higher job satisfaction (Clarke 2006). Only one study has specifically investigated the association between safety factors and nurse job satisfaction, but its analysis only controlled for one confounder (hospital size) (Barling et al. 2003), thus minimising its potential internal and external validity. More research is needed to examine whether hazard management would be a relevant strategy to enhance nurses’ job satisfaction.
Safety practices: need to assesses multiple factors and actors
According to Amick et al. (2000), safety practice refers to the extent to which workers perceive that safety and hazard management are priorities at their workplace. Safety practices have been related to disability management and occupational injury (Dennerlein et al. 2012). Given that safety practices are embedded in an organisational context, it is fundamental to account how different workplace actors such as supervisors, groups and individuals perceive them (Elovainio et al. 2000, Zohar & Luria 2005, Zohar & Tenne-Gazit 2008, Huang et al. 2012). In that sense, aggregating individual reports of safety practices by groups, for example nurses who work at the same unit, can provide a more accurate description of the safety actual situation at the unit, above and beyond varying self-reports of the same issue (Huang et al. 2007). However, very few studies have contemplated a multilevel framework to inquire about the role of safety practices, and consequently, more research is needed regarding how individual nurses, groups and supervisors jointly perceive safety practices at their respective units (Christian et al. 2009).
Multilevel modelling is particularly suitable for nursing research because nurses are typically organised into units under the direction of a nursing supervisor, so individual-level outcomes might be related to unit-level aggregated experiences as well as supervisor-level characteristics. Despite the appropriateness of conducting multilevel research, systematic reviews on job satisfaction in nursing and other occupations has mostly relied on self-reports without much consideration for workplace contextual actors and factors (Lu et al. 2005). This omission is significant because nurses’ health and well-being have been associated with team-aggregated assessments of the psychosocial work environment (e.g. shift control, meal break support) (Hurtado et al. 2015a,b). Consequently, less is known if psychosocial workplace experiences such as job satisfaction are also sensitive to supervisors’ assessments of their work environment. For instance, a study among manufacturing workers reported that job satisfaction was higher when employees perceived that their supervisor was responsive to their safety concerns (Ayim Gyekye & Salminen 2007). These analyses, however, relied on what workers reported about their supervisors lacking actual assessments from supervisors.
Unit supervisors merit attention because they not only are typically responsible for overseeing patient care within units but also have a critical role in safety management (Lin et al. 2011). Managerial and leadership roles have a positive impact on workplace safety, particularly if supervisors are perceived as receptive and responsive to workers’ concerns (Kelloway & Barling 2010). Consequently, evaluating how nursing supervisors appraise workplace safety is important given their central managerial role, particularly if safety-related factors are also likely to be associated with the job satisfaction of their staff (Duffield et al. 2009).
Study aims and hypotheses
This study aimed to address scientific gaps related to lacking multilevel evidence examining the associations between job satisfaction and safety practices as reported by nurses and their respective supervisors. Two hypotheses were tested: first, that assessments of safety practices would not differ between supervisors and their nurses (e.g. aggregated responses) (Hypothesis 1). Second, it was tested whether the job satisfaction of a given nurse would be higher if her unit-mates or unit supervisors reported safer practices (Hypothesis 2).
Methods
Participants
Design and context of the study
This was a cross-sectional study among hospital nurses and unit supervisors at two large academic hospitals in Boston, MA (USA).
Inclusion/exclusion criteria
Eligible participants included unit supervisors and registered nurses (RNs) working 20 work hours/week or designated as a minimum 0.5 full-time equivalent. Excluded workers were patient care assistants, those assigned to environmental services and physical medicine units (e.g. physical therapy, occupational therapy), those on an extended absence greater than 12 weeks and those deemed travelling or contract nurses.
Sample
Eligible participants were invited via e-mail to complete an online survey. E-mail reminders were sent after the fourth, sixth and 10th weeks, and paper versions of the survey were mailed to the non-respondents. A total of 1103 of RNs completed at least half of the online or paper survey (response rate of 79%). Likewise, 89 of 94 unit supervisors completed the survey (response rate of 84%). In this study, however, analyses were limited to nurses who provided information for all of the covariates (complete case analyses), for a total of 1052 RNs and 79 Supervisors in 79 units. All workers provided signed informed consent, and appropriate Institutional Review Boards approved methods and all project materials.
Measures
Response variable
Each nurse assessed job satisfaction with a widely used single item (Mangione & Quinn 1975, Lu et al. 2005): ‘all in all, how satisfied would you say you are with your job?’ Survey participants selected from a five-point ordinal response scale, from 1 (not at all satisfied) to 5 (extremely satisfied).
Predictor variable
Safety practices were measured with a subscale from the Organisational Policies and Practices questionnaire, an instrument developed to address organisational factors leading to injury and disability (Amick et al. 2000). The safety practices subscale encompassed five items measured on a five-point ordinal scale (1: strongly disagree, 5: strongly agree), regarding nurses’ assessments of the identification and improvement of unsafe conditions, e.g. ‘employees on this unit are trained in safe work practices for the job hazards they will encounter’. Individual responses of safety practices were aggregated (arithmetic mean) for all RNs at the same unit. Safety practices were also measured independently for each unit supervisor. The subscale had very high internal reliability coefficients for both supervisors and nurses, with corresponding Cronbach’s alphas of 0.83 and 8.11.
Control variables
We controlled for nurses’ individual-level characteristics. Socio-demographic and socio-economic covariates included age (18–24, 25–34, 35–44, 45–54, ≥55 years), race/ethnicity (non-Hispanic white vs. other), educational attainment (<12 years, 1–3 years of college or technical school, 4-year college degree, graduate school or more) and difficulty in paying bills as a proxy for income situation (1: great deal, 2: some, 3: a little, 4: not at all). Occupational covariates included shift work (working only on daytime or not), and hours worked per week (<30, 30–34, 35–39, 40–44, ≥45 hours).
Statistical analyses
Safety practices were measured with questionnaires applied for both nurses and their respective unit supervisors. The aggregation of individual responses by unit was performed after examining intra-unit correlations (e.g. the proportion of shared variance of safety practices of nurses at the same unit). Differences in safety practices between nurses and supervisors were examined with a paired t-test, and a Pearson correlation (Hypothesis 1). Though job satisfaction was measured with an ordinal, five-point Likert scale, diagnostic statistics showed a skewed positive distribution (e.g. only four nurses answered not being satisfied at all). Hence, the lowest two response options (not at all and not too satisfied) were merged into one category. As a result, the associations between safety practices with job satisfaction (Hypothesis 2) were tested logistic regression models in three steps; in the first step, unit-aggregated and supervisors’ safety practices scores were examined separately (a model for each predictor), adjusting only for individual-level confounders. The second step expanded the prior regression models, this time controlling for individual-level assessments of safety practices to account for potential residual or cross-level confounding. In the third step, the previous regression models were expanded including supervisors’ assessments of safety practices together. All analyses were two-tailed with a 0.05 level of statistical significance.
Results
Table 1 shows the distribution of socio-demographic and occupational characteristics of the sample. Among 1052 nurses, 92.6% were female and 89.1% were non-Hispanic white. Most nurses did not have a second job (85.2%) and worked regularly during the day shift (73.1%). Most supervisors were women (92%), over 50 years of age (66%) and with more than 4 years of tenure at their current units (45%) (Table 2).
Table 1.
Descriptive statistics for registered nurses (n = 1052)
| Characteristics | N (%) |
|---|---|
| Sex | |
| Male | 78 (7.4) |
| Female | 974 (92.6) |
| Age (years) | |
| 18–24 | 52 (4.9) |
| 25–34 | 288 (27.4) |
| 35–44 | 260 (24.7) |
| 45–54 | 291 (27.7) |
| ≥55 | 161 (15.3) |
| Race | |
| Hispanic | 22 (2.1) |
| White, non-Hispanic | 937 (89.1) |
| Black, non-Hispanic | 43 (4.1) |
| Other | 50 (4.8) |
| Education | |
| 1–3 years of college | 187 (17.8) |
| Baccalaureate degree | 650 (61.8) |
| Graduate degree | 215 (20.4) |
| Difficulty in paying bills | |
| Great deal of difficulty | 30 (2.9) |
| Some difficulty | 182 (17.3) |
| A little difficulty | 224 (21.3) |
| No difficulty | 582 (55.3) |
| Refused | 34 (3.2) |
| Hours worked per week | |
| <30 | 263 (25.0) |
| 30–34 | 135 (12.8) |
| 35–39 | 348 (33.1) |
| 40–44 | 265 (25.2) |
| ≥45 | 41 (3.9) |
| Having a second job | |
| Yes | 156 (14.8) |
| No | 896 (85.2) |
| Having only day shift, % | |
| Yes | 769 (73.1) |
| No (also night shift) | 283 (26.9) |
Table 2.
Descriptive statistics for supervisors (n = 79)
| Characteristics | N (%) |
|---|---|
| Age (years) | |
| 29–39 | 9 (12.0) |
| 40–49 | 16 (21.3) |
| 50–59 | 38 (51.7) |
| 60–66 | 12 (16.0) |
| Sex | |
| Female | 69 (92.0) |
| Male | 6 (8.0) |
| Race | |
| White | 75 (98.7) |
| Black | 1 (1.3) |
| Education | |
| Baccalaureate degree | 2 (2.6) |
| Graduate degree | 75 (97.4) |
| Tenure years at unit | |
| 1 | 3 (3.8) |
| 2 | 16 (20.3) |
| 3 | 24 (30.4) |
| 4 | 15 (19.0) |
| 5 | 21 (26.6) |
Differences in supervisors’ and nurses’ assessments of safety practices
Results from a t-test paired by units showed that supervisors assessed safety practices more favourably than their unit staff (Hypothesis 1). In a Likert scale from 1 to 5, supervisors’ average of safety practices was 4.02, whereas nurses’ aggregated average was 3.74, and this difference was statistically significance (P < 0.001). Moreover, supervisors’ and nurses’ aggregated scores of safety practices were not correlated (r = 0.10, P = 0.4).
Multilevel associations of safety practices with job satisfaction
The intra-unit correlation for safety practices was 0.11, meaning that about a tenth of the individual-level variance of this variable was attributable to the unit clustering. Table 3 shows the summary of the regression models that examined the association of job satisfaction with nurses’ aggregated responses and supervisors’ safety practices. Model I shows that nurses’ aggregated safety practices were positively associated with job satisfaction, independently of individual-level confounders. However, the association between job satisfaction and staff-aggregated scores of safety practices was attenuated after controlling for individuals’ assessments of safety practices (Model II), indicating the interference of residual confounding (i.e. differences in self-reports of nurses within each unit). This association remained statistically significant, and the adjusted odds of being in a higher category of job satisfaction were 1.67 times higher per one unit increment of safety practices. This association remained statistically significant after controlling for supervisors’ assessments of safety practices (Model II), which were not associated with job satisfaction, even when modelled independently of nurses’ aggregated responses (Model I).
Table 3.
Multilevel relationships of safety practices showing the odds ratio (OR) of nurses’ scoring a higher category of job satisfaction (n = 1052)
| Model I | Model II | Model III | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | OR | 95% CI | |
| Nurses’ aggregated safety practices | 4.69** | 3.08, 7.13 | 1.67* | 1.04, 2.68 | 1.65* | 1.02, 2.66 |
| Supervisors’ safety practices | 1.17 | 0.87, 1.58 | 1.09 | 0.85, 1.39 | 1.05 | 0.83, 1.34 |
P < 0.05,
P < 0.001.
Model I: Separate models for each predictor, adjusting for individual-level confounders including demographic (age, race, gender), socio-economic conditions (education level and capacity to pay the bills), occupational conditions (day shift or not, hours worked per week).
Model II: Expanded model I, adjusting also for individual-level assessments of safety practices.
Model III: Expanded model II, but both predictors were included in the model simultaneously.
Discussion
This study examined whether nurses and supervisors would differ in their assessments of safety practices, and if nurses’ personal job satisfaction would be higher if supervisors and unit-mates reported safer practices at their respective units. Supervisors’ safety practices were not associated with the job satisfaction of their staff. However, the odds of higher job satisfaction were 67% higher for nurses at units with better safer practices, compared with nurses at units that averaged lower safety practices. Supervisors’ and nurses’ safety assessments were not correlated, yet supervisors tended to assess the safety practices at their units more favourably than their supervised staff.
Further investigation on the sources of this difference would be important to understand fully why supervisors’ assessments were not associated with the job satisfaction of their staff. However, differences in safety practices could be explained by the fact that nurses’ aggregated assessments are more likely to reflect actual practices at their units, whereas supervisors’ assessments would mostly be based on formal procedures (Ossmann et al. 2005). Other supporting evidence has shown that supervisors may overestimate characteristics of the work environment, perceiving them more favourably than their supervised staff (Carson et al. 1998, Huang et al. 2012). These findings suggest the possibility that supervisors, despite their critical leadership role, may not always accurately assess the safety situation at their units.
Results are parallel to the conclusions reached by an extensive international study where less than half of surveyed nurses reported that nursing management was responsive to their concerns or that they were provided opportunities to participate in decision-making (Aiken et al. 2001). This study complements previous evidence showing how safety-related factors affect other relevant outcomes such as medical errors (Hofmann & Mark 2006). Salient safety practices include supervisors’ responsiveness and proactiveness to employees’ concerns (Kelloway & Barling 2010). Yet, these results showed that supervisors’ safety assessments and nurses’ job satisfaction are independent, despite evidence linking job satisfaction with supervisory leadership styles such as inspiring a shared vision (Chiok Foong Loke 2001), or supervisory practices such as patient-focused work allocation and accountability for patient care (Makinen et al. 2003). Future studies, then, should investigate more closely how different dimensions of organisational behaviours are related to nurses’ job satisfaction depending on the actors that report them.
Study limitations and strengths
This study has two main limitations. First, job satisfaction was measured with a widely used single-item question that, however, does not capture specific dimensions of job satisfaction related to working conditions, self-growth and promotion and job security (Lin et al. 2011). Second, because of the cross-sectional survey design, reverse causation cannot be ruled out, for example, if workers who were not satisfied with their job were more likely to report less safe practices. To circumvent this potential bias, regression models accounted for individual-level assessments; even though these assessments attenuated the association between unit-level nurse assessments of safety practice and job satisfaction, the magnitude of the coefficient was still statistically significant. In turn, this study also has several strengths including a large random sample of nurses with high response rates (which favours the external validity of the findings) and the inclusion of multiple control factors relevant to health care settings (Williams et al. 2007).
Implications for nursing management
Results suggest that safety practices might affect job satisfaction in a high-risk industry such as nursing. This study highlights the importance that nursing leaders such as directors of nursing or unit supervisors routinely calibrate their personal safety assessments with in-depth evaluations of their supervised staff. The risk of adverse occupational outcomes could be lower if nursing supervisors are proactive addressing safety concerns (Kelloway & Barling 2010), a practice that could also boost job satisfaction. This study is one of the few in nursing research to measure safety aspects at multiple levels, including assessments of supervisors and their staff. Results showed that job satisfaction was higher at units with safer practices. Supervisors’ safety assessments were not associated with the job satisfaction of their staff. Future studies are required to examine the role of safety practices in enhancing job satisfaction as well as supervisors’ roles and responsibilities in enhancing both safety practices and workers’ job satisfaction.
Acknowledgements
This work was supported by a grant from the National Institute for Occupational Safety and Health (NIOSH) to the Harvard School of Public Health Center for Work, Health and Well-being. This study would not have been accomplished without the participation of Partners HealthCare System and leadership from Dennis Colling, Sree Chaguturu and Kurt Westerman. The authors would like to thank Partners Occupational Health Services including Marlene Freeley for her guidance, as well as Karen Hopcia, Elizabeth Taylor, Elizabeth Tucker O’Day and Terry Orechia. We also thank individuals at each of the hospitals, including Jeanette Ives Erickson and Jacqueline Somerville in Patient Care Services leadership, and Jeff Davis and Julie Celano in Human Resources. Additionally, we wish to thank Charlene Feilteau, Mimi O’Connor, Margaret Shaw, Eddie Tan and Shari Weingarten for assistance with supporting databases. We also thank Chris Kenwood of NERI for his statistical and programming support, Julie Theron, Project Director, and Linnea Benson-Whelan for her assistance with the production of this manuscript. D.A.H. led the writing of the manuscript. S.-S.K. and G.S. conceptualised the study. S.-S.K. analysed the data and made the first draft. G.S., S.V.S., J.T.D., D.C.C. and D.M.H. provided the critical reviews for the draft. All authors participated in revising the paper.
Source of Funding
This work was supported by a grant from the National Institute for Occupational Safety and Health (U19 OH008861) for the Harvard T.H. Chan School of Public Health Center for Work, Health and Well-being.
Footnotes
Ethical Approval
Harvard School of Public Health Institutional Review Board
References
- Aiken LH, Clarke SP, Sloane DM et al. (2001) Nurses’ reports on hospital care in five countries. Health Affairs 20 (3), 43–53. [DOI] [PubMed] [Google Scholar]
- Aiken LH, Clarke SP, Sloane DM, Sochalski J & Silber JH (2002) Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. Journal of the American Medical Association 288 (16), 1987–1993. [DOI] [PubMed] [Google Scholar]
- Aiken LH, Sermeus W, Van den Heede K et al. (2012) Patient safety, satisfaction, and quality of hospital care: cross sectional surveys of nurses and patients in 12 countries in Europe and the United States. British Medical Journal 344, e1717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Amick BC, Habeck RV, Hunt A et al. (2000) Measuring the impact of organizational behaviors on work disability prevention and management. Journal of Occupational Rehabilitation 10, 21–38. [Google Scholar]
- Ayim Gyekye S & Salminen S (2007) Workplace safety perceptions and perceived organizational support: do supportive perceptions influence safety perceptions? International Journal of Occupational Safety and Ergonomics 13 (2), 189–200. [DOI] [PubMed] [Google Scholar]
- Barling J, Kelloway EK & Iverson RD (2003) High-quality work, job satisfaction, and occupational injuries. Journal of Applied Psychology 88 (2), 276. [DOI] [PubMed] [Google Scholar]
- Carson PP, Carson KD, Yallapragada R & Roe CW (1998) Nursing supervisors’ view through rose-colored glasses: an overestimation of employees’ positive attitudes. The Health Care Manager 16 (4), 68–74. [DOI] [PubMed] [Google Scholar]
- Chiok Foong Loke J (2001) Leadership behaviours: effects on job satisfaction, productivity and organizational commitment. Journal of Nursing Management 9 (4), 191–204. [DOI] [PubMed] [Google Scholar]
- Christian MS, Bradley JC, Wallace JC & Burke MJ (2009) Workplace safety: a meta-analysis of the roles of person and situation factors. Journal of Applied Psychology 94 (5), 1103. [DOI] [PubMed] [Google Scholar]
- Clarke S (2006) The relationship between safety climate and safety performance: a meta-analytic review. Journal of Occupational Health Psychology 11 (4), 315. [DOI] [PubMed] [Google Scholar]
- Dennerlein JT, Hopcia K, Sembajwe G et al. (2012) Ergonomic practices within patient care units are associated with musculoskeletal pain and limitations. American Journal of Industrial Medicine 55 (2), 107–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Duffield C, Roche M, O’Brien-Pallas L, Catling-Paull C & King M (2009) Staff satisfaction and retention and the role of the nursing unit manager. Collegian 16, 11–17. [DOI] [PubMed] [Google Scholar]
- Elovainio M, Kivimäki M, Steen N & Kalliomäki-Levanto T (2000) Organizational and individual factors affecting mental health and job satisfaction: a multilevel analysis of job control and personality. Journal of Occupational Health Psychology 5 (2), 269. [DOI] [PubMed] [Google Scholar]
- Gershon RR, Stone PW, Zeltser M, Faucett J, MacDavitt K & Chou SS (2007) Organizational climate and nurse health outcomes in the United States: a systematic review. Industrial Health 45 (5), 622–636. [DOI] [PubMed] [Google Scholar]
- Hofmann DA & Mark B (2006) An investigation of the relationship between safety climante and medication errors as well as other nurse and patient outcomes. Personnel Psychology 59 (4), 847–869. [Google Scholar]
- Huang Y-H, Chen J-C, DeArmond S, Cigularov K & Chen PY (2007) Roles of safety climate and shift work on perceived injury risk: a multi-level analysis. Accident Analysis & Prevention 39 (6), 1088–1096. [DOI] [PubMed] [Google Scholar]
- Huang YH, Verma SK, Chang WR et al. (2012) Supervisor vs. employee safety perceptions and association with future injury in US limited-service restaurant workers. Accident Analysis & Prevention 47, 45–51. [DOI] [PubMed] [Google Scholar]
- Hurtado DA, Glymour MM, Berkman LF, Hashimoto D, Reme SE & Sorensen G (2015a) Schedule control and mental health: the relevance of coworkers’ reports. Community, Work & Family 18 (4), 416–434. [Google Scholar]
- Hurtado DA, Nelson CC, Hashimoto D & Sorensen G (2015b) Supervisors’ support for Nurses’ meal breaks and mental health. Workplace Health & Safety 63 (3), 107–115. [DOI] [PubMed] [Google Scholar]
- Kelloway EK & Barling J (2010) Leadership development as an intervention in occupational health psychology. Work & Stress 24 (3), 260–279. [Google Scholar]
- Lin BY, Hsu CP, Juan CW, Lin CC, Lin HJ & Chen JC (2011) The role of leader behaviors in hospital-based emergency departments’ unit performance and employee work satisfaction. Social Science & Medicine 72 (2), 238–246. [DOI] [PubMed] [Google Scholar]
- Lu H, While AE & Barriball KL (2005) Job satisfaction among nurses: a literature review. International journal of Nursing Studies 42 (2), 211–227. [DOI] [PubMed] [Google Scholar]
- Makinen A, Kivimaki M, Elovainio M, Virtanen M & Bond S (2003) Organization of nursing care as a determinant of job satisfaction among hospital nurses. Journal of Nursing Management 11 (5), 299–306. [DOI] [PubMed] [Google Scholar]
- Mangione TW & Quinn RP (1975) Job satisfaction, counter-productive behavior, and drug use at work. Journal of Applied Psychology 60, 114. [DOI] [PubMed] [Google Scholar]
- Ossmann J, Amick BC 3rd, Habeck RV et al. (2005) Management and employee agreement on reports of organizational policies and practices important in return to work following carpal tunnel surgery. Journal of Occupational Rehabilitation 15, 17–26. [DOI] [PubMed] [Google Scholar]
- Utriainen K & Kyngas H (2009) Hospital nurses’ job satisfaction: a literature review. Journal of Nursing Management 17 (8), 1002–1010. [DOI] [PubMed] [Google Scholar]
- Williams RM, Westmorland MG, Shannon HS & Amick BC 3rd (2007) Disability management practices in Ontario health care workplaces. Journal of Occupational Rehabilitation 17, 153–165. [DOI] [PubMed] [Google Scholar]
- Zohar D & Luria G (2005) A multilevel model of safety climate: cross-level relationships between organization and group-level climates. Journal of Applied Psychology 90 (4), 616. [DOI] [PubMed] [Google Scholar]
- Zohar D & Tenne-Gazit O (2008) Transformational leadership and group interaction as climate antecedents: a social network analysis. Journal of Applied Psychology 93 (4), 744–757. [DOI] [PubMed] [Google Scholar]