This was a cross-sectional study conducted among Bangladeshi registered nurses available on online platforms and the nurses working in eight tertiary level hospitals of two large administrative divisions (Dhaka and Sylhet) in Bangladesh. These two administrative divisions were conveniently selected because Dhaka is the capital and Sylhet is a significant region of Bangladesh. The self-reported data were collected between 26 February 2021 and 10 July 2021.

The study population of this study was all registered nurses of Bangladesh working in clinical settings for at least 1 year. Our required sample size was 1,024 at 80% power, 95% CI of 0.05 to 1.96, and 3% margin of error with an assumption that 50% of the nurses had symptoms of burnout. Finally, from 1,345 obtained responses, 1,264 completed responses were considered in the analysis. Thus, an additional 11% of participants (240) helped to reduce the study’s margin of error [34].

An online and offline method of data collection was approached in this study. A semi-structured self-response questionnaire was developed to gather data for this study’s findings. The questionnaire had five parts. On the front page, study’s objectives and the responding procedure were described. The first part of the questionnaire consisted of the demographic and occupational characteristics of the respondents. In the subsequent parts of the questionnaire, workplace bullying, violence, and burnout-related questions were documented with proper instructions for responding to them. Finally, three dichotomous-response (yes/no) questions related to participants’ “presence of enjoyment in the current job,” “presence of enthusiasm in the current job,” and “presence of satisfaction in the current job” were asked. The questionnaire was translated into Bangla with the help of an expert. One nurse superintendent of a tertiary hospital and one public health expert in Bangladesh reviewed the initial questionnaire. Based on their suggestions, modification of the questionnaire was performed. For online data collection, convenience and snowball-sampling methods were followed. At the time of the pandemic, face-to-face data collection was restricted. So, a questionnaire link (using “Google Form”) was distributed on common social media platforms used by nurses (Facebook, WhatsApp, etc.) in Bangladesh, and available registered nurses were invited to participate. By online method of data collection, 721 completed responses were obtained. To achieve the required sample size, we distributed another 700 printed questionnaires in eight hospitals in two geographical divisions (Dhaka and Sylhet) in Bangladesh. The respondents were provided with 7 days to respond to the questionnaire, and data collectors received questionnaires after 7 days. After receiving 655 returned copies, 543 were obtained as completed responses. Thus, a total of 1,264 completed responses were finally included in the current study.

The Short Negative Acts Questionnaire [S-NAQ] was used to determine workplace bullying exposure [35]. It comprises nine items that assess whether or not a person has been subjected to bullying behaviors in the last 6 months. The scale items address both the personal and work-related forms of bullying (e.g., “there has been gossip or rumors spread about you” and “necessary information was withheld that impeded your ability to do your job”). The answer categories ranged from 1 to 5, with 1 being “never” and 5 being “daily”. The S-NAQ has a good level of reliability and validity [35]. This scale has been used to measure bullying in numerous studies in several countries, including Belgium, Italy, and Norway [36–38]. The 9-item S-NAQ has a Cronbach's α of 0.89 in our current study. S-NAQ has two cut-off scores: 15 and 23, which means that respondents scoring less than 15 in the S-NAQ can be considered “non-exposed” to bullying at work; those scoring between 15 and 22 are at “high risk” of being bullying victims or may be immersed in a bullying process; whereas those scoring 23 or higher can be considered “targeted” of workplace bullying [39].

The workplace violence scale (WVS) was used to measure the WPV experienced in the last 12 months [40, 41]. Several studies used this scale to measure workplace violence among nurses and other health care workers previously [42–45]. The scale was composed of five types of WPV named physical assault, emotional abuse, threat, verbal and sexual harassment, and sexual assault. The responses (score ranges from 0 to 3) of the five items scale indicate the frequencies of each type of WPV. Score 0 indicates WPV 0 time/none, 1 indicates 1 time, 2 indicates 2 or 3 times, and 3 indicates WPV more than 4 times. By summing all the responses, the total score ranged from 0 to 15. Four WPV categories were derived from scores 0 for none, 1 to 5 for low, 6 to 10 for intermediate, and 11 to 15 for high. In the questionnaire, participants were given specific definitions of each type of violence. Cronbach’s α reliability coefficient was calculated at 0.60, which indicated an acceptable internal consistency of the scale.

The 10-item Burnout Measure-Short version (BMS) developed by Malach-Pines was used in the last segment of our survey [46]. The BMS is a brief and easy-to-use tool to measure burnout. In this tool, an individual’s levels of physical exhaustion, emotional exhaustion, and mental exhaustion are assessed using the core elements of the concept of burnout in a series of 10 questions. Each item is graded on a 7-point Likert scale from 1 (never) to 7 (always). For all 10 items, the total response points ranged from 10 to 70, based on the response value for each item. The overall burnout score for each participant was calculated by dividing the sum of each participant’s response values by 10. Thus, the overall burnout score ranged from 1 to 7. An overall score ≥4 indicates an established state of burnout, according to Malach-Pines. Therefore, individuals are divided into two categories based on their overall burnout score: those who are likely to experience burnout (overall score ≥4) and those who are not likely to experience burnout (overall score <4). The 10-item BMS was validated and used on different samples and has shown satisfactory psychometric properties [46–50]. In our study, the Cronbach’s α for the 10-item BMS was 0.89, indicating good reliability.

The demographic profile and occupational characteristics of the study sample were described using descriptive statistics expressed in frequency and percentages. A chi-square test was performed for unadjusted associations. Mixed-effects Poisson regression models were fitted to find the adjusted association between burnout and WPB, WPV, and other study variables. The factors in the unadjusted test were included in the adjusted models at a priori specified p-value of 0.1. Mixed-effects Poisson regression models with robust error variance were used to avoid overestimation of associations with common binary outcomes measured in cross-sectional studies [51–53]. We fitted three models to investigate the adjusted association between the predictor variables and the outcome variable. Model 1 included only demographic variables. Model 2 included both demographic and occupational variables. Model 3 included all main-effects terms (demographic and occupational variables) and two-way interaction-effects terms between WPB and WPV. The findings were expressed as relative risks (RRs) with 95% confidence intervals (CIs). Moreover, the association between burnout and three job-related questions was evaluated using chi-square test. All statistical analyses were two-sided, and a p-value < 0.05 was considered statistically significant. Statistical software STATA-16 (Stata Corp LP, College Station, TX, United States) was used for data analysis.

The ethical review committee of Begum Rabeya Khatun Chowdhury Nursing College, Bangladesh (approval ID: BRKCNC-IRB-2021/5) approved this study involving human participants. On the first page of the questionnaire, the study’s aims and objectives were explained. The participation of respondents in the study defined their implied consent.

The characteristics of the 1,264 participants are presented in Table 1. The mean age of the participants was 28.41 (SD: 5.54), and 70.02% were female. The 756 (59.81%) nurses were involved in government hospitals, and the remaining 508 (40.19%) were in private hospitals. Moreover, 910 (71.99%) of the nurses were from tertiary care hospitals. More than half of the nurses (52.93%) reported that they did not have enough equipment to manage the patients while on the job. A substantial percentage of respondents (79.27%) stated that they had never been awarded for good work, and 81.80% of nurses stated that they had never been trained against WPV.

Burnout was found to be prevalent in 685 (54.19%) of the 1264 nurses. Among the nurses, 781 (61.79%) reported workplace bullying (WPB), with 529 (41.85%) and 252 (19.94%) nurses reported “high risk bullying” and “targeted bullying,” respectively. According to the survey, 206 nurses (16.30%) reported “intermediate and high” levels of workplace violence during the last 12 months (Table 1).

Table 2 represents the burnout distribution as well as the unadjusted relationship between burnout and WPB, WPV, and other factors. In the chi-square test, WPB was found to be significantly associated with burnout of nurses (p < 0.001). Similarly, WPV was significantly associated with nurses’ burnout (p < 0.001). Almost half of the female nurses (49.27%) were found to be burnout. In demographic variables, sex, geographical division of workplace, and educational degree of the nurses were found to be significantly associated with burnout (p < 0.001). The occupational variables that were significantly associated with burnout were job experience of the nurses, timely salary, working hours, sleeping hours, number of patients dealt with per day, the sufficiency of the equipment to manage patients, rewards for good job performance, and training for the nurses against WPV (p < 0.001).

Table 3 represents the adjusted association of workplace bullying, burnout, demographic, and occupational factors with burnout identified from three mixed-effects Poisson regression models. According to Model 3, among the nurses, “high risk bullying” and “targeted bullying” were at 2.29 (95% CI: 1.53–3.41, p < 0.001) and 4.86 (95% CI: 3.32–7.11, p < 0.001) times more risk of burnout, respectively, compared to non-exposed groups to the bullying. The “low” and “intermediate and high” levels of WPV groups were 1.48 (95% CI: 1.03–2.11, p < 0.033) and 3.65 (95% CI: 2.40–5.56, p < 0.001) times more risk of being burnout, respectively, compared to WPV non-exposed groups. In terms of their association with burnout, WPB and WPV were found to interact with each other, as two combinations of WPB and WPV were significant (high risk bullying × intermediate and high WPV, p = 0.002 and targeted bullying × intermediate and high WPV, p < 0.001) compared with their respective reference combinations. Females were at 1.35 (95% CI: 1.19–1.53, p < 0.001) times more risk of burnout. Compared to the nurses who completed diploma educational degrees, the nurses who completed bachelor’s and master’s degrees were at 1.21 (95% CI: 1.05–1.40, p = 0.010) and 1.14 (95% CI: 1.00–1.28, p = 0.042) times more risk of being burnout, respectively. Less experienced (<3 years) nurses were at 1.33 (95% CI: 1.19–1.49, p < 0.001) times more risk of being burnout compared to the nurses with ≥3 years of job experience. Nurses with higher working hours (>48 h per week) were at 1.22 (95% CI: 1.04–1.42, p = 0.014) times more risk of being burnout than the nurses who worked ≤36 h per week. The nurses who served the highest number of patients (≥40 patients per day) were at 1.20 (95% CI: 1.05–1.38, p = 0.008) times more risk of being burnout than the nurses who served <20 patients per day. The nurses who usually did not get rewards from their authority for good job performance were at 1.33 (95% CI: 1.13–1.58, p = 0.001) times more risk of burnout than their counterparts. Nurses who did not receive any training against WPV were at 1.21 (95% CI: 1.02–1.44, p = 0.031) times more risk of burnout than their opposite parts.

We asked nurses three job-related questions. Compared to the nurses who suffered burnout, non-burnout nurses found enjoyment in job (46.18% vs 69.46%, p < 0.001), were enthusiastic (44.02% vs 67.25%, p < 0.001), and satisfied (42.11% vs. 65.35%, p < 0.001) with their present job (Table 4).

Nurses are the most important staff in a country’s healthcare system as they deal with the health and well-being of the patients. We anticipate that our study findings could help the authorities, the Directorate General of Health Services, the Ministry of Health and Family Welfare, and the Directorate General of Nursing and Midwifery, Bangladesh to understand the workplace environment of nurses and take the necessary measures to reduce the frequency of workplace bullying and violence.

To our best knowledge, this is the first study that addressed Bangladeshi nurses’ burnout status and its associated factors, including the association with WPB and WPV. Nurses from all geographical divisions of the country got opportunities to participate in the study. The larger sample size might provide substantial accuracy to the study findings. However, this research had some limitations also. As a non-random sampling technique was applied, selection bias could not be excluded. The risk of information bias might be present due to the self-reported questionnaire. As a nature of a cross-sectional study, causality could not be established. However, we hope that the outcomes of our research will provide a succinct summary of the working environment of Bangladeshi nurses. Further in-depth and rigorous research is recommended focusing on nurses’ burnout, WPB, and WPV to establish sustained evidence to make nurses’ working environments safer.

A high prevalence of burnout, workplace violence, and bullying was found among Bangladeshi nurses. Workplace violence and bullying were also identified as potential predictors of burnout. Burnout is known to have a negative impact on job satisfaction, which leads to a higher likelihood of turnover in the workplace. As a result, establishing a safer working environment is critical, as is demanding improved nursing services. Thus, addressing the variables, enhancing the working environment for nurses, increasing job satisfaction, as well as lowering burnout, bullying, and workplace violence are crucial through support and the implementation of suitable policies from hospital administrators, policymakers, and the government.