Abstract
BACKGROUND:
The limited research has focused on the relationship between work-related stress and self-efficacy in relation to mental health problems in nurses.
OBJECTIVE:
This multi-hospital cross-sectional survey investigated the relationships between work-related stress, self-efficacy and mental health status of hospital nurses in Tabriz, Iran.
METHODS:
Four hundred hospital nurses completed a questionnaire including demographic and job details, Health & Safety Executive (HSE) Management Standards Revised Indicator Tool (MS–RIT), General Self-Efficacy (GSE–10) scale, and General Health Questionnaire (GHQ–28). A three-step hierarchical logistic regression modelling was used.
RESULTS:
Work-related stress, self-efficacy and mental health problems were significantly related to each other. The results of the regression modelling revealed that working overtime and number of patients cared for were significant positive predictors, while job tenure as well as control and relationships dimensions of work stress were significant negative predictors of mental health problems, with the final model explaining 21% of the variance in the outcome measure. Addition of self-efficacy at step 3 did not result in a significant change in the variance from previous steps.
CONCLUSIONS:
The results provide further support for stress prevention strategies focused at the job (e.g., better organisation of work demands) and organisational (e.g., improving employee participation and involvement in work) levels.
Introduction
Work-related stress arises from interactions between employees, work environments, and cultural contexts [1]. Work-related stress is a major challenge in both developed and developing countries [2–4] and the percentage of employees experiencing this problem has been increasing over time [1]. Work-related stress has serious implications for both individuals (e.g. in terms of health and performance, coping strategies, job satisfaction, family and social adjustments) and organisations (e.g. increased staff turnover, intention to leave and education/training costs to improve staff morale) [4–9]. Work-related stress occurs when expectations of an individual go beyond his/her capabilities. It can generally be considered as harmful physical and psychological responses in an individual resulting from lack of balance between job demands and an individual’s abilities [10]. These psychosocial hazards that can result in work-related stress underpin both the predominant causation models and the development of interventions to address work-related stress [11, 12].
Nursing personnel may experience a relatively high level of stress [6, 7] that is often attributable to a variety of potent workplace stressors such as conflict with physicians, discrimination, high workload, shift working, low social support, nursing shortage, lack of adequate attention to nursing profession, bullying and violence, and dealing with death and dying issues [3, 13–15]. Moreover, continuous organisational changes such as increased demands, over-expectations of patients, development of new technologies and therapeutic methods which are inseparable elements of health systems, are also considered as other important psychosocial factors contributing to stress in this working population [16]. Work-related stress factors affect the mental, physical, and social well-being of nurses [5, 17–19], and have been shown to be associated with negative consequences such as job dissatisfaction, burnout, turnover, intention to leave, absenteeism and decreased professional/organisational commitment, productivity, work ability and performance [3, 21]. These undesired consequences and work ability and performance decrements resulting from work-related stress in nurses may negatively affect the quality and quantity of patient care [3, 22]. Thus, understanding the characteristics of workplace stress factors has important implications in terms of both nurses’ health and patient outcomes.
Perceived self-efficacy is another cognitive factor which is of particular importance in the nursing profession, considering the fact that these employees are faced with job stressors on a regular basis. Self-efficacy can be defined as: “people’s judgements of their capabilities to organise and execute courses of action required to attain designated types of performances” [23]. The existence of various stressors in the work environment may have significant impact on nurses’ beliefs or confidence, and consequently on their health and performance. However, to the authors’ knowledge, limited research has focused on the relationship between work-related stress and self-efficacy in relation to mental health problems in nurses. In this regard, two recent studies evaluated the relationship between stress and self-efficacy in nurses [24, 25], and only one of them focused specifically on job stress dimensions and reported negative association between stress and self-efficacy [24]. In addition, the causes and levels of work-related stress as well as its effects vary greatly depending on the work context and cultural background [7, 26]. Thus, it seems reasonable to assume that levels and patterns of job stress experienced by nurses vary worldwide depending on their work settings and levels of social support. Moreover, the need for better understanding of the effects of individual variations in perceptions and reactions to stressors on health outcomes in nurses is well emphasised [7, 27]. Finally, there is still more to learn from different organisational and cultural contexts about the combined effects of work stress and self-efficacy on mental health status of nurses. It is, therefore, important to understand better the most critical aspects of workplace stress in nursing occupations, and to develop techniques and activities aiming to promote their mental health.
In an attempt to address the above-mentioned issues, the present work was carried out to evaluate the levels of work-related stress, self-efficacy, and their relationship with mental health status of hospital nurses.
Methods
Study design, procedure and participants
A multi-hospital cross-sectional descriptive-analytical study was performed in eight teaching hospitals in Tabriz, Iran. Permission to conduct the study was obtained from the hospital authorities involved and the research ethics committee at the Tabriz University of Medical Sciences approved the study. Participation was voluntary, all information was kept strictly confidential, and participants had the right to refuse participation. Before the study commenced, all participating nurses were familiarised with the study conditions and signed an informed consent form. The study population was hospital nurses in the eight teaching hospitals. Having an associate degree or higher in nursing, working for > 1 year in their job, and having no history of mental disorders were considered as inclusion criteria for the study. The selected hospitals had a total number of 1908 nurses at the time of the study. A two-stage sampling method was applied for selection of participants. First, the study participants were randomly selected using a probability proportion to size sampling method. A random numbers table was then used to select nurses from each unit. Four hundred hospital nurses (329 females and 71 males) declared their agreement to assist in the study.
Data collection
Information on demographic and job characteristics, work-related stress, perceived self-efficacy, and mental health status were collected using a questionnaire. Instruments used in the study included: (1) Health & Safety Executive (HSE) Management Standards Revised Indicator Tool (MS–RIT), (2) General Health Questionnaire (GHQ–28), and (3) General Self-Efficacy (GSE–10) scale. Demographic and job details including gender, age, educational level (undergraduate, postgraduate), marital status (single, married), job tenure, number of hours worked per day, shift working (no, yes), working overtime (no, yes), and number of patients cared for were also recorded.
The 35-item MS–RIT [28], with established reliability and validity [2], was used to assess workplace stress. Advantages of this instrument include its short length and ability to measure multiple dimensions of workplace stress [28]. It consists of seven subscales including demands (e.g., work procedures, work load and work environment –eight items), control (e.g., control and influence over work –six items), managerial support (e.g., support and information from superiors –five items), peer support (e.g., support and encouragement from colleagues –four items), relationships (e.g., conflict and unacceptable behaviour at work –four items), role (e.g., clear job roles, expectations, responsibilities and goals –five items) and change (e.g., management and communication of changes in the organisation –three items). Response to the questions of this instrument is based on two 5-point Likert scales (‘always’ to ‘never’ and ‘strongly disagree’ to ‘strongly agree’). The items for demands and relationships subscales were negatively phrased, which were reversed for comparison purposes. The total score for the MS–RIT ranges from 35–175. Higher scores on this instrument indicate low risk of stress, while lower scores suggest high risk of stress.
The 28-item GHQ [29] was used for evaluation of the mental health status (e.g., mental distress). This is a well-tried and tested technique, which has four subscales including: somatic symptoms (seven items), anxiety symptoms and sleep disorders (seven items), social dysfunction (seven items), and depression symptoms (seven items). Each item is measured on a 4-point scale ranging from ‘never true’ = 0 to ‘always true’ = 3. The total scores for each subscale and for the GHQ-28 range from 0–21 and 0–84, respectively. Higher scores on this scale indicate more severe mental problems. Both the total score and sub-dimensions of the GHQ can be used for the purpose of analysis. In this study, the total score of the GHQ was used as a measure of mental health problems for analytical purposes [14, 30]. In addition, a cut-off point of 23 [31] was used for description of cases with poor mental health.
The 10-item GSE scale was used to evaluate the self-efficacy [32]. This is a one-dimensional scale and items are measured on a 4-point scale (‘totally incorrect’ = 1 to ‘totally correct’ = 4). The total score for the GSE ranges from 10–40. Higher scores on this scale show higher levels of self-efficacy.
Data analysis
The SPSS software version 21.0 (SPSS Inc., Chicago, IL, USA) was used for data analysis. Descriptive statistics (means and standard deviations [SD]), reliability coefficients, and bivariate correlations were calculated for the study variables. Multivariate assumptions were examined and fulfilled prior to statistical analyses. A three-step hierarchical regression was used to comprehensively examine the relationships among self-efficacy, work-related stress, and mental health problems. For this, independent variables were categorised into three levels. In step 1, control variables including gender (female = 0, male = 1), age, job tenure, shift working (no = 0, yes = 1), working overtime (no = 0, yes = 1), and number of patients cared for were added to the model. This was followed by addition of the work-related stress (dimensions of MS–RIT) in step 2 and self-efficacy (GSE-10 score) in step 3. Standardised regression coefficients (β) and explanatory power (adjusted R-square –R2) were calculated. P values < 0.05 were considered statistically significant.
Validity and reliability
Valid and reliable Farsi (Iranian language) versions of the MS–RIT [33], GHQ–28 [31], and GSE–10 [34] were used. The Cronbach’s alpha for work stress dimensions ranged from 0.70 to 0.89, and for the whole scale was 0.79. The Cronbach’s alpha for the GHQ–28 subscales ranged from 0.69 to 0.88, and for the whole scale was 0.80. Finally, the Cronbach’s alpha for the GSE–10 was 0.90.
Results
Characteristics of the sample
Mean age and job tenure were 35.7 years (SD = 6.6 years; range = 20 to 57 years) and 10.9 years (SD =4.9 years; range = 1 to 30 years), respectively. Most of participating nurses were females (n = 329; 82.3%), married (n = 256; 64.0%), shift worker (n = 349; 87.3%), had undergraduate education (n = 342; 85.5%), and worked overtime (n = 298; 74.5%). More than half of the participants worked more than 8 hours per day (n = 212; 53.0%) and cared for less than 10 patients per day (n = 234; 58.5%).
Mean (SD) score for the MS-RIT was 109.4 (11.6). Mean scores for dimensions of work stress ranged from 9.1 (SD = 2.1) for the change dimension to 25.7 (SD = 4.3) for the demand dimension (Table 1). Mean total GHQ–28 score was 22.3 (SD = 10.2; range = 3 to 61). The social dysfunction (mean = 7.8; SD = 3.4) had the highest and depression symptoms (mean = 3.3; SD = 2.6) had the lowest mean scores. The proportion of nurses scored above the cut-off point (GHQ-28 > 27.38) was 31.3%. Mean GSE–10 score was 17.5 (SD = 5.0; range = 10 to 40).
Descriptive statistics, Cronbach’s alphas, and Pearson’s correlation coefficients of the study variables
Descriptive statistics, Cronbach’s alphas, and Pearson’s correlation coefficients of the study variables
*Correlation is significant at the 0.05 level (2-tailed); **Correlation is significant at the 0.01 level (2-tailed).
The results of correlation analysis (shown in Table 1) indicated that work-related stress, self-efficacy and mental health problems were significantly related to each other. The Pearson correlation coefficients ranged from 0.12 to 0.61.
Hierarchical regression analysis
The results of hierarchical regression analysis are presented in Table 2. In step 1, job tenure and working overtime were significant predictors of mental health problems among studied nurses. However, variables in this step did not account for any significant variance in mental health problems, R2 = .04, F(6, 393) = 3.97, p < 0.001. In step 2, stress dimensions explained 16% of the variance in mental health problems, F (13, 386) = 9.12, p < 0.001. Control (β= –0.26, p < 0.001) and relationships (β= –0.33, p < 0.001) dimensions of work stress were significant negative predictors of mental health problems among nurses. Addition of self-efficacy at Step 3 did not result in a significant change in variance from previous steps, R2 = 0.21, F(14, 385) = 8.56, p < 0.001. The final regression model explained 21% of the variance in mental health problems.
Summary of hierarchical regression analysis for variables predicting mental health problems
Summary of hierarchical regression analysis for variables predicting mental health problems
Statistically significant values are shown in bold. †Adjusted R2. ‡R2 change. *p < 0.05, **p < 0.01, ***p < 0.001.
The influence of demographic and job characteristics, self-efficacy, and work-related stress on mental health of hospital nurses were investigated. It was shown that the work-related stress and mental health problems were common among the participants, which is similar to reports from other countries and is a reflection of stressful work conditions of the study participants. Within the overall model, working overtime and number of patients cared for were significantly positive predictors of mental health problems, while job tenure as well as control and relationships dimensions of work stress were significant negative predictors of mental health problems in the studied nurses.
Our findings regarding workplace stress are consistent with previous research, which have shown that job stress is fairly common in nursing population [7, 36]. However, there are also studies which have shown a relatively higher level of work stress among nurses [37, 38] compared to other professions. Such a difference may be attributed to the use of different tools for evaluation of stress in this working population. In addition, work-related stress may also depend to a large extent on the work setting. For instance, Tajvar et al. showed high levels of occupational stress among intensive care unit nurses, while our participants were selected from both general and intensive care units [38]. It is acknowledged that working in intensive care units imposes a high level of stress on nurses due to high workload, unit space, and exposure to suffering patients or patient death [39]. It is therefore not surprising to observe that the level of experienced job stress and its multiple dimensions are different in each working environment. These findings underscore the importance of studying workplace stress and its multiple dimensions in each work setting.
The results from this study identified underlying work stress dimensions among hospital nurses that can be used to develop appropriate strategies to address this problem. The scores of the MS–RIT subscales indicated that all dimensions of work stress, except for the role dimension, had significant contribution to the experience of stress among hospital nurses. Therefore, primary workplace interventions aiming at improving work organisation (e.g., job demands and control as well as managerial and social support), and through developing strategies to cope with changes in task demands are recommended with a view to helping prevent work stress in these employees. To take account of this, different stress reduction techniques such as coping strategies, better work organization, decreasing workloads, sharing more time with family and friends, changing communication techniques, and managing personal expectations and attitudes [1, 40] can be utilized to decrease work-related stress among nurses.
The study results also showed a relatively high level of mental distress among hospital nurses. This is in agreement with some previous studies [2, 42], which have reported relatively high levels of mental health problems among nurses. The study finding that social dysfunction was more prevalent than other aspects of mental distress in the studied nurses highlights the need for increased attention to this aspect of mental health in this working group.
In addition, our findings showed significant relationship between work stress and mental health problems, which is not surprising in view of previous research [3, 38]. More specifically, control and relationships aspects of work stress were predictors of mental health problems in the final regression model. In addition to this, job factors including job tenure, working overtime and number of patients cared for were also predictive of mental health problems in the final model. Bazazan et al. also reported that job experience was negatively associated with mental health problems in nurses [43]. Taken together, the above findings emphasise the need for consideration of various sources of stress at work. The implications of these findings for nursing practice are to: 1) design, allocate, and organise work demands based on the employees’ capabilities and limitations, 2) use of participatory ergonomic approaches in order to give employees a greater sense of control and influence over work tasks and decisions, and 3) promote positive working relationships and improve workplace morale. This is of particular interest, since emphasis on improving stressors in the work environment is not only essential for health and well-being of employees, but is also important in terms of quality of care and hospital costs.
However, our findings provide evidence that self-efficacy is not a significant predictor of mental health problems in hospital nurses. In other words, addition of self-efficacy did not change the variance from previous steps. Thus, it can be concluded that self-efficacy did not moderate the relationship of work-related stress and mental health in this working population. This finding is not in agreement with the general assumption that employees with high level of self-efficacy have higher perception of control and that control is likely to moderate the relationship of stress and well-being [44]. Similarly, the results confirmed that the demographic variables (e.g., gender and age) were not predictors of mental health problems. Nevertheless, Milutinović et al. found that socio-demographic factors such as age and marital status influenced work stress perception [39], which is not similar to our results. Others have suggested an integrated stress prevention strategy at both individual and organisational levels [3]. Nevertheless, our findings provide more support for the stress prevention strategies involving job and organisational factors.
The findings should be interpreted in the light of some limitations such as generalisability and study design (e.g., cross-sectional). However, as most health care employees are exposed to some extent to workplace stressors, these findings are likely to be applicable to other health care professionals. Nevertheless, it may be useful to replicate this research to improve understanding and generalisability of these findings.
Conclusions
This study provides further evidence that workplace stress, low self-efficacy, and mental distress are significant problems in hospital nurses, and provides a basis for intervention programmes to improve the health and well-being of this group of healthcare professionals. Within the overall regression model, working overtime, job tenure, and number of patients cared for as well as control and relationships dimensions of work stress were significant predictors of mental health problems, while self-efficacy had no effect. It can be concluded that stress prevention strategies should be focused at the job (e.g., better organisation of work demands) and organisational (e.g., improving employee participation and involvement in the work) levels.
Footnotes
Acknowledgments
The authors wish to acknowledge the participants and those who facilitated the research in the study hospitals.
Funding
This research received no funding from any agency.
Conflict of interest
The authors declare no conflicts of interest.
