Study design, settings, and participants
We conducted a cross-sectional study among Bangladeshi registered male nurses between April 26 and July 10, 2021. By assuming a 50% prevalence of TI among male nurses, a required sample size of 384 was determined based on a 95% confidence level and a 5% margin of error [30]. After the data collection and cleaning, a total of 379 male nurses were included in this study. The criteria of the study participants were directly involved in clinical care settings, registered nurse (RN) according to the Bangladesh Nursing and Midwifery Council (BNMC) [31], and had at least one year of work experience. The study sites included indoor or outdoor settings where nurses provide healthcare, such as private and non-private tertiary hospitals, secondary hospitals, upazila hospitals, community clinics, and other healthcare sites.
Study questionnaire
The questionnaire used in this study incorporated both the original English version and a translated version in Bengali to ensure a comprehensive understanding among respondents. The translation process involved two independent translators converting the questionnaire from English to Bengali. The translated version was carefully compared to the original English version by the authors, and any ambiguities or discrepancies were discussed with the translators to refine the Bengali version of the questionnaire. Face validation was conducted to enhance the validity and clarity of the questionnaire. Five nurse superintendents and five psychologists were involved in reviewing the questionnaire. Their valuable input, comments, and suggestions regarding the wording and layout were carefully considered. Based on the suggestions, slight modifications were made to the wording, meanings, and content of each item in the questionnaire. Following the refinement process, a pilot test was conducted among a group of 20 nurses to evaluate the suitability and comprehensibility of the revised questionnaire. This pilot testing helped identify any potential issues or challenges that needed to be addressed before the main data collection. The questionnaire consisted of three sections. The first section focused on collecting demographic information from the participants. The second section aimed to gather occupational characteristics. Finally, the third section contained items specifically related to WPB and TI measurement.
Study variables
The outcome variable of our study was the TI, while the main exposure variable was the WPB experienced by male nurses. After the literature review, we included demographic variables, including age, place of residence, educational degree, and smoking status, as well as work-related variables such as job type, professional titles, monthly salary, timely payments, availability of accommodation facilities, presence of sufficient equipment to perform work, rewards for good works, and training against workplace violence which was reported to be associated with the nurse TI [32–42].
WPB measurement
To evaluate the workplace bullying (WPB) experienced by nurses, we utilized the "Short Negative Acts Questionnaire (S-NAQ-9)", which consists of nine items [43]. This questionnaire was employed to determine whether individuals have been subjected to WPB behaviours within the preceding six months. The scale encompasses various forms of personal and WPB, such as "there has been gossip or rumors spread about you", "necessary information was withheld that impeded your ability to do your job." Respondents were presented with response options ranging from 1 to 5, where 1 represented "never" and 5 represented "daily." Therefore, the overall score on the scale ranged from 9 to 45, with higher scores indicating a greater degree of WPB experienced. The internal consistency of the S-NAQ-9, as measured by ω (McDonald's omega), was determined to be 0.88 in the current sample, indicating good internal reliability.
TI measurement
We employed the Turnover Intention Scale-6 (TIS-6) to measure the TI among the nurses over a period of nine months. This scale, consisting of six items, has been widely utilized in prior studies focusing on nurses [44, 45]. Participants were instructed to respond on a five-point Likert scale, ranging from 1 to 5, to indicate their agreement with statements pertaining to the impact of their current job on their intention to leave, the frequency at which they consider alternative job options, and the extent of hesitation they experience when contemplating leaving their present position. A composite score ranging from 6 to 30 was calculated by summing the responses to all items, with higher scores indicating a greater degree of TI. The reliability coefficient of the scale, as determined by ω, was found to be 0.89.
Data collection procedure
The data collection process employed a convenience sampling method using the semi-structured self-reported questionnaire. We utilized both online and offline approaches to collect the data. For online data collection, eligible participants were invited to participate in the study by accessing an online questionnaire link, which was facilitated using "Google Form." This link was shared across various social media groups, such as Facebook, WhatsApp, and other relevant platforms, to reach out to nurses. Nurses willing to participate clicked on the provided link to complete the questionnaire. In addition to online data collection, in-person data collection was carried out by distributing printed copies of the questionnaire among male nurses in selected tertiary hospitals in two major divisions of Bangladesh ("Dhaka and Sylhet"). We allowed participants seven days to fill out the questionnaire, and after that, the research assistants collected completed questionnaires. After the participants filled out the questionnaires, the collected data were manually entered into the same Excel sheet to ensure uniformity and consistency. By employing these combined methods, a total of 420 responses were collected, and subsequently, 379 completed responses were included in the analysis.
Data Analysis
The data analysis and management process involved the utilization of Stata 16 and Microsoft Excel 2016. Microsoft Excel was employed to compile the data obtained from both offline and online data collection methods. Subsequently, the compiled data sheet was imported into STATA for further analysis. Descriptive statistics, including percentages and frequencies, were calculated for categorical variables, while measures such as mean, standard deviation, minimum, and maximum were computed for continuous variables. Inferential statistical techniques, such as the t-test and one-way analysis of variance (ANOVA), were employed to assess mean differences in TI across different demographic and occupational characteristics. Pearson's correlation test was conducted to examine the relationships between age, monthly income, WPB, and TI. Multiple linear regression models were fitted to examine the adjusted association between WPB and TI while controlling for demographic and occupational characteristics. Three hierarchical regression models were developed to identify predictors. Model 1 included demographic variables, Model 2 incorporated occupational variables in addition to demographic variables, and Model 3 encompassed WPB, along with both demographic and occupational variables. The final multiple hierarchical regression models encompassed WPB, demographic variables, and occupational variables to explore the combined influence of the study variables. To assess multicollinearity, a post-estimation test known as the Variance Inflation Factor (VIF) was performed. In this study, statistical significance was determined using a threshold of p < 0.05.
Ethical considerations
Ethical considerations were duly addressed in our study by the Ethical Committee of Begum Rabeya Khatun Chowdhury Nursing College, Bangladesh, with approval ID BRKCNC-IRB-2021/5. Participants were asked to provide their informed consent before participating in the study. Participants were provided with the freedom to refuse participation, and they were informed of their right to withdraw from the study at any time. We ensured the confidentiality of participants' data and provided a comprehensive briefing on the aim and objectives of the study.