Evaluating effort-reward imbalance among nurses in emergency departments: a cross-sectional study in China

Our study was part of a large nationwide study on the health and working conditions of ED staff in China, which was conducted from July to August 2018. This cross-sectional study was an online survey undertaken in coordination with the Medical Administration Bureau of the National Health Commission of the People’s Republic of China. In hospitals providing pre-hospital care in China, this questionnaire was distributed on the ED nurses’ work platform and administrated by an online survey platform (platform name: Questionnaire Star, website: https://​www.​wjx.​cn). Finally, the questionnaire link was clicked by 25,518 ED nurses, who were thought to be invited into our study. Of them, 17,582 nurses completed the online questionnaire, with a response rate of 68.9%.

This study was approved by the Research Ethics Committee of Hainan Medical University, Haikou, China (protocol number: HYLL-2018-035). All the methods in this study were in accordance with the institutional research committee and tenets of the Declaration of Helsinki. All participants provided informed consent. Moreover, each individual participated in this anonymous survey voluntarily, and their information was kept confidential.

Data were collected using a structured anonymous questionnaire developed based on a review of the published literature [5‐10]. A total of 30 ED nurses participated in a preliminary test in Wuhan. Based on the pre-test, the questionnaire was modified to ensure that the questions were clear and understandable. In this study, the questionnaire content included three parts: sociodemographic characteristics, work-related factors, and ERI. Sociodemographic characteristics and work-related factors were selected based on published literature in this field [5‐9]. The ERI questionnaire used in this study was widely applied internationally, and its Chinese version has been proven to have good reliability and validity [10, 19].

The sociodemographic characteristics included sex, age, marital status and educational level. The work-related factors included professional title, hospital level, monthly income, years of service, frequency of night shifts, perceived shortage of nurses, frequency of workplace verbal violence and experiencing workplace physical violence in the past year. The perceived shortage of nurses was assessed by one question: ‘Do you think that the current number of nurses in the ED can meet daily work needs?’ (‘Does not meet demands’, ‘General’, ‘Meets demands’).

The balance between job efforts and rewards was evaluated by an effort-reward ratio (ERR), which was measured using subscales of the ERI questionnaire [10, 19]. Accordingly, six items for job efforts, including time pressure, work interruption, responsibility, extra work, physical labour, and workload increase, were rated on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Thus, the total score of the measured efforts ranged between 6 and 30. The higher the score, the greater the perceived effort. Eleven items for job rewards, including esteem, job promotion and job security, were rated on a five-point Likert scale ranging from 1 (strongly agree) to 5 (strongly disagree). Thus, the total score of the measured rewards ranged from 11 to 55. The higher the score, the more the perceived rewards were experienced. The ERR was calculated according to the following formula: ERR = (11 × effort)/(6 × reward). Therefore, an ERR value greater than 1.0 indicated a high level of ERI, which reflected that the invested efforts were not matched by the rewards received. In this study, the Cronbach’s alpha values of efforts and rewards were 0.82 and 0.92, respectively.

Pearson’s chi-squared test was used to conduct descriptive analyses of sociodemographic characteristics and work-related factors in different effort-reward status groups. Categorical variables were reported as frequencies and percentages. Moreover, the associations of the independent variables with ERI were explored using binary logistic regression and a stepwise selection method. The odds ratios (ORs) and 95% confidence intervals (CIs) of the variables were reported. Spearman’s rho correlations and the variance inflation factor (VIF) were used to test for multicollinearity between independent variables. The Spearman’s rho correlations were at the maximum threshold of 0.7 and the VIF was at maximum threshold of 10 (mean VIF = 1.33, max VIF = 2.07 and minimum = 1.06), which indicated that no collinearity was detected (Supplementary Table 1). A Hosmer-Lemeshow goodness-of-fit test was used to examine model fitness (p = 0.309). All the analyses were conducted at a statistical significance level of 0.05, using the Statistical Analysis System (SAS) version 9.4 for Windows (SAS Institute Inc., Cary, NC, USA).

The basic characteristics of the ED nurses are summarised in Table 1. A total of 17,582 ED nurses completed the survey; 10.25% of them were male and 89.75% were female. The ages of the ED nurses tended to be young and 79.81% were younger than 35. In this sample, the educational level was mainly an associate degree, which accounted for 48.93% of the participants. Correspondingly, 61.81% of them were married. Moreover, in this survey, 37.33%, 39.55%, and 4.14% of the ED nurses worked in level III, level II, level I hospitals respectively, and 18.99% worked in emergency centres. In addition, Table 1 describes the different characteristics of ED nurses between the effort-reward balance and ERI groups. In this study, 59.66% of the ED nurses experienced ERI. Consequently, significant differences were found between the groups for the following items: sex, age, marital status, educational level, professional title, hospital level, years of service, perceived shortage of nurses, frequency of night shifts, workplace verbal violence and workplace physical violence (p < 0.05).

Table 2 presents the relationships between ERI and sociodemographic and work-related factors. The risk of experiencing ERI was higher among ED nurses aged 25 to 34 than among those aged less than 25 (OR = 1.13, 95% CI: 1.03–1.24). Our results also revealed that when ED nurses were male (OR = 1.24, 95% CI: 1.11–1.39), had a bachelor’s degree or above (OR = 1.21, 95% CI: 1.05–1.39), junior professional title (OR = 1.13, 95% CI:1.02–1.25), and intermediate or above professional title (OR = 1.43, 95% CI: 1.24–1.65), they had higher odds of experiencing ERI. ED nurses with a higher monthly income were less likely to exhibit ERI (OR = 0.73, 95% CI: 0.64–0.82). Reverse trends were observed for years of service and the frequency of night shifts. ED nurses who had longer years of service (OR = 1.48, 95% CI: 1.28–1.71) and had a higher frequency of night shift (OR = 1.93, 95% CI: 1.61–2.31) were more likely to experience ERI. Compared to perceiving a shortage of nurses, when the number of nurses met the demands of the department, the risk of ERI among them markedly decreased (OR = 0.32, 95% CI: 0.30–0.35). In addition, compared to not having suffered verbal violence in the last year, the rate of the occurrence of ERI increased 0.64 times and 1.72 times when ED nurses had suffered workplace verbal violence one to three times (OR = 1.64, 95% CI: 1.52–1.76) and four times or more (OR = 2.72, 95% CI: 2.48–2.99), respectively. Analogously, ED nurses who had suffered workplace physical violence in the last year were 1.32 times more likely to experience ERI than those who had not (OR = 1.32, 95% CI: 1.19–1.48).

This was a national cross-sectional study with a large sample size. To the best of our knowledge, this study is the first to investigate the ERI and its influencing factors among ED nurses in China. In addition, there were some limitations to the findings. First, our study focused only on ED nurses, and our findings should be carefully generalised to other populations. Second, it was based on convenience sampling, which may have a selection bias. It is also expected that nurses who were more interested in the study subject were more likely to participate in the survey, which may have resulted in an overestimation of ERI levels. However, as this study was a nationwide and large sample survey, the sample should be representative to some extent. Third, it is not possible to draw a causal relationship because of the cross-sectional nature of the study. Therefore, further prospective studies are required.