SEARCH

SEARCH BY CITATION

Keywords:

  • Medical ward(s);
  • nurses;
  • psychiatric hospital;
  • work-related stress

Abstract

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Purpose

Little is known about the level of work-related stress in nurses in China. This study compared the level of work-related stress between female nurses working in psychiatric and general hospitals in China.

Design and Methods

A descriptive comparative cross-sectional design was used. A consecutive sample of nurses from two psychiatric hospitals (N = 297) and a medical unit (N = 408) of a general hospital completed a written survey including socio-demographic data and a measure of work-related stress (Nurse Stress Inventory).

Findings

Compared to the nurses working in the general hospital, those working in the psychiatric setting had a higher level of stress in the domains of working environment and resources (p < .001) and patient care (p < .001), but lower workload and time (p < .001). Multivariate analyses revealed that college or higher level of education (β = .1, p < .001), exposure to violence in the past 6 months (β = .2, p < .001), longer working experience, and working in psychiatric hospitals were associated with high work-related stress (β = .2, p < .001).

Practice Implications

Considering the harmful effects of work-related stress, specific stress management workshops and effective staff supportive initiatives for Chinese nurses are warranted.

Work-related stress (WRS) is defined as a harmful emotional and somatic response when the person's skills and resources cannot satisfy the requirement of the task (Leka, Hassard, & Yanagida, 2012). WRS leads to considerable negative effects for both individuals and society. For example, in the United Kingdom 13.4 million sick days were related to stress, anxiety, and depression in 2001 with cost estimates of approximately £4 billion/year (Gray, 2000; Health and Safety Executive, 2005). Of all the occupations, nurses are one of the most at risk for WRS (Sherring & Knight, 2009).

A large number of studies on stress in nurses have been conducted in Western clinical settings. The common sources of stress at work identified include shift work, long working hours, lack of control and conflicting demands, bad relations with colleagues, low pay, and poor working environments (Edwards et al., 2006; Kawano, 2008; Toh, Ang, & Devi, 2012; UNISON, 2002). Considering that cultural factors can play an important role in stress, it is important to be cautious when applying results obtained in Western societies to different cultures (Yau et al., 2012). In addition to cultural differences, there are also local effects on WRS. For example, even nurses working in different hospital units are exposed to different types of WRS (Kawano, 2008).

In the past years, several studies have examined WRS in nurses in China. Yau et al. (2012) interviewed 93 nurses in a general hospital and reported that nurses had a relatively high level of stress, particularly in the domains of working environment and resources, and workload and time as measured by the Nurse Stress Inventory (NSI) (Li & Liu, 2000). Xianyu and Lambert (2006) examined a convenience sample of 92 senior nurses from two hospitals and found that sources of stress included heavy workload, interpersonal conflict, and low nurse–patient ratio. Common limitations of these studies include the small sample size and lack of multivariate analyses exploring independent factors contributing to higher level of stress. In addition, given the more specific features in psychiatric hospitals than general clinical settings, such as restricted wards and higher risk of violence toward nurses, to date there are no studies comparing WRS in nurses between these two settings in China.

The purpose of this study is to (a) compare the level of WRS between nurses working in two major psychiatric hospitals and a major general hospital in China, and (b) explore the association of individual and work environment attributes with WRS.

Method

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Design

After approval by the Clinical Research and Ethics Committee of the respective hospitals, a descriptive comparative cross-sectional survey was conducted between March 1 and 31, 2012.

Study Participants and Settings

The setting was two major psychiatric hospitals and the medical department of a major general hospital in Beijing, China. All female Registered Nurses currently working directly with patients were invited to participate in this study. Those who worked less than 6 months in the hospitals were excluded. Unlike in psychiatric hospitals, there are very few male nurses working in general hospitals in China. In order to make the two samples homogenous and comparable, only female nurses were included in this study, and only the Medical Unit of a general hospital was included.

Measures

A self-report structured questionnaire designed for this study was used to collect data on basic demographic characteristics and exposure of workplace violence against nursing staff in the past 6 months. Following the literature (Shiao et al., 2010), workplace violence in the past 6 months used in this study included the following items: sexual assault with violence (the experience of unwelcome sexual advances involving violence), sexual harassment with physical contacts (as the experience of any type of unwelcome sexual behavior with body contact), verbal harassment with sexual content (the experience of verbal abuse with sexual content), nonsexual physical violence (the experience of being subjected to physical contact, such as being hit, kicked, slapped, pushed, grabbed, choked, etc., with intention to harm or injure), and verbal threat and abuse (being sworn at, yelled, called by names or other words intended to control or hurt) by patients or their families (Shiao et al., 2010). Respondents responded to these items with “yes” or “no.” If they answered “yes” to any of these items, they were classified as “experiencing workplace violence.”

WRS was measured by the 35-item NSI (Li & Liu, 2000) that comprises five domains: (a) nursing profession and clinical duty (social status, continuing education, wages and benefits, opportunities for promotion, frequency of shift duty, and support from management), (b) workload and time (heavy workload, number of coworkers, psychological counseling for patients, non-nursing workload, and amount of paperwork), (c) working environment and resources (poor working environment, lack of necessary equipment, and overcrowded wards), (d) patient care (accidents at work, recognition of nursing work by patients and their families, patients' demand, cooperation and relationship with patients and their families, level of proficiency of nursing, and the impact of patients' suffering and sudden death), and (e) management and interpersonal relationship (lack of understanding and respect from others, excessive criticism from nurse managers and doctors, lack of understanding and support between colleagues, conflict with the doctors and nurse managers, and lack of friendship and cooperation between colleagues). The Cronbach coefficients of internal consistency reliability ranged from .83 to .98 for all domains (Li & Liu, 2000).

The NSI was based on the Nurse Stress Scale (Gray-Toft & Anderson, 1981) and the Source of Stress Inventory (Wheeler & Riding, 1994), and has been widely used in recent studies (An, Wang, & Wang, 2005; Yau et al., 2012). A 5-point Likert scale (0 = never, 1 = seldom, 2 = sometimes, 3 = often, 4 = always) was used with higher scores indicating higher level of stress. The mean score of each stress domain was calculated.

Data Collection

The questionnaires were personally provided to all female nurses by the managers of the Nursing Departments and then collected in a sealed box within 2 weeks. A letter introducing the purpose of the survey and assuring confidentiality and anonymity was attached to the questionnaires. All participating nurses signed an informed consent form if they agreed to participate in this study.

Statistical Analysis

The data were analyzed with SPSS 13.0 for Windows (SPSS Inc., Chicago, IL, USA). The comparisons between psychiatric and medical nurses' demographic characteristics and WRS were performed using independent sample t test, Mann–Whitney U test, and chi-square test, as appropriate (Table 1). Comparisons between the above two groups in each domain of WRS were further conducted by analysis of covariance with age, marital status, education, shift pattern, clinical working experience, exposure to workplace violence in the past 6 months, monthly income, and job rank as the covariates. The independent contributors of each domain of WRS were examined by stepwise multiple linear regression analyses; each domain of WRS was the separate dependent variable, while age, marital status, education, shift pattern, working experiences, exposure to workplace violence in the past 6 months, monthly income, and job rank were entered as the independent variables. The normality of distributions for the continuous variables was evaluated using the one-sample Kolmogorov–Smirnov test. Two-tailed tests were used in all the analyses with the significance level set at .01 due to multiple comparisons.

Table 1. Demographic Characteristics of Nurses by Work Location (Psychiatric and General Hospitals)
 The Whole Sample (n = 705)Medical Nurses (n = 408)Psychiatric Nurses (n = 297)Statistics
N%N%N%χ2dfp
  1. a

    Mann–Whitney U test.

Married/cohabitating41158.323357.117859.9.61.5
College education or above18626.411528.27123.91.61.2
Work shift      25.61<.001
Fixed day11916.94410.87525.3   
Rotating58683.136489.222274.7   
Being senior rank11215.95012.36220.99.61.002
Workplace violence in the past 6 months33547.59623.523980.5223.51<.001
 MSDMSDMSDT/Zdfp
Age (years)31.37.130.66.532.27.86−3.0703.002
Monthly income (CNY)4,0764,3013,5111,9204,8536,156−7.6a<.001
Working experiences (years)10.47.89. 47.211.88.4−4.1703<.001
Work-related stress         
Nursing profession and clinical duty2.6.72.6.82.6.7.3703.8
Workload and time2.7.82.8.82.5.86.0703<.001
Working environment and resources2.31.02.11.02.5.9−5.9703<.001
Patient care2.3.52.2.42.5.6−9.7703<.001
Management and interpersonal relationship2.1.82.2.92.1.7.8703.4

Results

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The response rate was 99.2% (705 returned of 711 questionnaires distributed). Table 1 compares the basic demographic factors and WRS of psychiatric and medical nurses. Compared to medical nurses, psychiatric nurses were more likely to work on fixed day shifts (10.8% vs. 25.3%, p < .001), exposed to more workplace violence (23.5% vs. 80.5%, p < .001), had an older age (30.6 vs. 32.2 years, p = .002), had a higher monthly income (3,511 vs. 4,853, p < .001) and longer working experience (9.4 vs. 11.8 years, p < .001), and had a higher level of stress in the domains of working environment and resources (p < .001) and patient care (p < .001), but lower level of stress in the domain of workload and time (p < .001).

After controlling for the potential confounders, the significant differences between the two groups in workload and time (F(1,698) = 40.3, p < .001), working environment and resources (F(1,698) = 13.7, p < .001), and patient care (F(1,698) = 37.5, p < .001) remained, whereas there was still no difference in nursing profession and clinical duty (F(1,698) = 3.6, p = .06) and management and interpersonal relationship domains (F(1,698) = 2.6, p = .1).

Results of stepwise multiple regression analysis with each domain of WRS as dependent variables and demographic and clinical factors as predictors are shown in Table 2. College or higher level of education was independently associated with high level of stress in the domains of nursing profession and clinical duty (β = .1, p < .001), workload and time, working environment and resources (β = .1, p < .001), and patient care (β = .1, p = .001). Exposure to violence in the past 6 months contributed to higher WRS in the domains of nursing profession and clinical duty (β = .2; p < .001), workload and time (β = .1, p = .005), and patient care (β = .2; p < .001). Longer working experiences contributed to higher WRS in the domains of workload and time (β = .2; p < .001) and working environment and resources (β = .1; p = .001). Working as a nurse in a psychiatric hospital contributed to high level of WRS in working environment and resources (β = .2; p < .001) and patient care (β = .3; p < .001), but low level of WRS in workload and time (β = −.3; p < .001).

Table 2. Stepwise Multiple Regression Analysis With Domains of Work-Related Stress as Dependent Variables and Demographic and Clinical Factors as Predictors
Dependent VariablePredictorβT valuep
  1. Only statistically significant predictors are reported.

Nursing profession and clinical duty

Adjusted R2 = .04;

F(3, 701) = 13.5; p < .001

College education or above.13.7<.001
Workplace violence in the past 6 months.23.9<.001

Workload and time

Adjusted R2 = .10;

F(5, 699) = 17.1; p < .001

Psychiatric hospitals−.3−6.9<.001
Work experiences.24.8<.001
College education or above.13.8<.001
Workplace violence in the past 6 months.12.8.005

Working environment and resources

Adjusted R2 = .08;

F(4, 700) = 16.2; p < .001

Psychiatric hospitals.23.5<.001
Work experiences.13.2.001
College education or above.13.4.001

Patient care

Adjusted R2 = .15;

F(3, 701) = 43.7; p < .001

Psychiatric hospitals.35.9<.001
College education or above.13.5.001
Workplace violence in the past 6 months.24.2<.001

Discussion

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Medical and psychiatric nurses experience different levels of WRS. Compared to the nurses working in the medical department, those working in the psychiatric setting had a higher level of stress in the domains of working environment and resources and patient care, but lower workload and time. Multivariate analyses revealed that college or higher level of education, exposure to violence in the past 6 months, longer working experience, and working in psychiatric hospitals were associated with high WRS.

In this study, psychiatric nurses had higher level of stress in domains of working environment and recourses and patient care than the nurses working in general hospital. There are several possible reasons that could explain this finding. First, all psychiatric nurses in this study worked in restricted environments, such as closed wards with isolation rooms that mainly serve acute and potentially violent patients. A considerable proportion of these patients displayed aggressive behavior previously and were compulsively admitted. Therefore, they were exposed to psychological and physical stress and aggression at work more frequently than nurses in general care setting (Krakowski, Kunz, Czobor, & Volavka, 1993). Second, relative to nonpsychiatric medical facilities, there is a strong need for continuing education and training about workplace violence, particularly how to support each other and being sensitive to subtle signs that can indicate proneness to aggressive behavior. However, this type of training is not universally implemented or inadequate in Chinese psychiatric hospitals. For example, a survey of 300 randomly selected nurses from five psychiatric hospitals in China found that 75.6% were not satisfied with the lack of continuing training on violence (Liu, Hou, Tian, Hu, & Li, 2011).

Medical nurses had higher levels of stress in the domains of workload and time than psychiatric nurses. In China, many people prefer to seek health care in large general hospitals even if they do not have serious diseases. These large urban hospitals by nature have higher turnover of inpatients and a large number of outpatient visits in these medical settings (Xie, Want, & Chen, 2011). Because of the higher patient volume, nurses' workloads are heavy and they experience a higher level of stress in this domain than their counterparts in psychiatric hospitals.

Consistent with previous findings (Fujita et al., 2012; Gates, Gillespie, & Succop, 2011; Littlejohn, 2012; Shiao et al., 2010; Whittington, Shuttleworth, & Hill, 1996), workplace violence was linked to higher level of stress in the domains of workload and time, working environment and resources, and patient care. College level or higher education was associated with high level of stress in all domains, a finding that was not reported previously. Unlike their Western counterparts, Chinese nurses have relatively low social status; nurses often complain of being verbally and physically attacked by patients and their families, and even being yelled at by administrative staff and doctors (Lambert, Lambert, Petrini, Li, & Zhang, 2007). Therefore, we assume that the gap between the low social status and the high expectations of the nurses with higher degree may account for the association between higher education and WRS.

Nurses with longer work experience are more likely to have developed clinical skills to deal with complex clinical situations. Therefore, it is safe to assume that longer work experience would be associated with lower level of WRS. Yet, having longer work experience was linked to higher level of stress in domains of workload and time and working environment and resources in this study. There are two possible reasons for this association. First, experienced nurses are more likely to face difficult patients when their junior colleagues cannot deal with them, which may lead to overwhelming WRS. Second, the three hospitals included in this study were teaching hospitals. Thus, in addition to clinical work, experienced nurses were more frequently responsible for professional education and research in contrast to their junior colleagues, which also increased their WRS.

There are three limitations to consider when interpreting these results. First, because of the cross-sectional design of this survey the causality of relationship between WRS and its correlates could not be explored. Second, only three major hospitals in China were involved, and the results may not be generalized to all clinical settings in China. Third, certain data, such as workplace violence, were collected retrospectively, which might induce recall bias. Fourth, some factors with the potential of influencing WRS, such as management styles in the two selected hospitals, were not measured in the study.

In conclusion, this study supports that nurses working in psychiatric and general hospitals were exposed to different types of WRS. Given the negative impact of stress on both individuals and the institutions, specific and culturally sensitive interventions in reducing WRS for psychiatric and medical nurses should be developed in China.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The study was supported in part by grants from the National Science Foundation of China (No. 30800367; 30770776; 81171270) and the Beijing Nova Program of the Beijing Municipal Science and Technology Commission (No. 2008B59).

References

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References