The prevalence of job stressors among nurses in private in vitro fertilization (IVF) centres

Abstract Aim The primary aim of this study was to identify the level of stress and the stressors having an impact on nurses compared with other medical workers in private IVF centres. Background Stressful working conditions can an adversely affect not only the health and well‐being of health professionals but also subsequently to patient outcomes if care is given to infertile couples. This is of relevance particularly in view of Vietnam's recent economic growth and the increase in the number of private IVF centres. This is the first study looking at the levels of stress experienced by health workers (especially nurses) providing IVF services. Design A cross‐sectional survey. Methods All health workers in seven IVF Clinics in HCMC were invited to complete an Occupational Stress Index (OSI) questionnaire. Results Of the invited 131 medical professionals, 105 (80%) completed the confidential self‐administered questionnaire. Thirty‐five participants (33.3%) were nurses, 19 (18.1%) were doctors and 51 (48.6%) were lab technicians. Approximately two‐thirds reported not having children (67.6%), half (50.48%) married and three‐quarters (76.2%) were women, with a significant difference by medical worker group (p < 0.05). Among the three groups, nurses have higher occupational stress index score compared with the others. The OSI score only had a strong relationship with the “high demand” (p < 0.001). Some demographic variables (e.g., income, long working hours, education level) statistically represented the high significant source of job stress.

In addition, working in very stressful environments with minimal control and social support from colleagues may also have a negative effect on patient safety (Berland, Natvig, & Gundersen, 2008).
Moreover, in vitro fertilization (IVF) centres are known to be particularly stressful (Costantini-Ferrando, Joseph-Sohan, Grill, Rauch, & Spandorfer, 2016;Harata et al., 2012). The high emotional content of consultations involved in the management of infertile couples is a contributory factor to the stress (Greenfeld, 1997;Oddens, den Tonkelaar, & Nieuwenhuyse, 1999). The unpredictable outcome of the treatment is another major stress-inducing agent, more likely to evoke feelings of depression (Dunkel-Schetter & Lobel, 1991). Stressful working conditions can adversely affect not only the health and well-being of health professionals but also subsequently to patient outcomes (Halm et al., 2005;Leiter, Harvie, & Frizzell, 1998;Vahey, Aiken, Sloane, Clarke, & Vargas, 2004) if care is being given to infertile couples. This is of relevance particularly in view of Vietnam's recent economic growth and the increase in the number of private IVF centres. To our knowledge, this is the first study looking at the levels of stress experienced by health workers (especially nurses) providing IVF services.

| Aims
The primary aim of this study was to identify the level of stress and the stressors having an impact on nurses compared with other medical workers (including physicians and IVF laboratory technicians) in private IVF centres to inform the development of strategies to benefit staff well-being and quality of care for patients.

| Design
A cross-sectional study based on the survey design.

| Participants
From August 2016 to November 2016, all employees involved in seven private IVF centres in South Vietnam were invited to participate in the study. Only permanent employees of these IVF centres were included. Interns, medical students, residence, collaborating staffs and university-employed clinical lecturers were excluded.

| Data collection
Information about the study was included with the invitation letter and those who were willing to participate completed an Occupational Stress Index (OSI) questionnaire. Completion of the questionnaire was deemed to signify consent to participate in the study. The medical workers themselves actually filled out the three specific OSI questionnaires, such as nurses-specific OSI, physician-specific OSI and generic OSI, tailored respectively to nurse, physician and IVF laboratory technician. OSI questionnaires were completed by participants without the intervention of the researchers after the participants were fully informed about the aims and methods of the study. These questionnaires were usually conducted at lunchtime or after working hours. The OSI of every participant was given a unique number on the questionnaire. Principle Investigator (PI) kept a list of names and form numbers. The PI attended again the following week and collected completed OSI by recruiting late starters and any new medical workers. Then gentle reminders to complete were sent to participants by SMS. Research assistants chased up late forms on behalf of PI.

| Ethical considerations
The study was reviewed by Scientific Board of Research Center for Genetics and Reproductive Health (CGRH)-National University and was conducted after the Ethics Committee approval of My Duc Hospital had been granted.

| Data analysis
The Occupation Stressor Index (Belkic, 2000;Belkic & Nedic, 2007Levi et al., 2000;Nedic, Belkic, Filipovic, & Jocic, 2010), a twodimensional matrix, was defined by the vertical axis composed of levels of information transmission (including input, central decisionmaking, output, general) and the stressor aspects placed along the horizontal axis. Each element in the OSI was scored from 0 to 2 (0: "not present," 2: "strongly present") and summed to give total scores under each domain. The domains are underload, high demand, strictness, external time pressure, aversive physical exposures, symbolic aversiveness (or avoidance) and conflict (or uncertainty). Summed scores for the aspects and the total OSI were calculated according to this framework (Belkić & Savić, 2013).
The data was stored in Microsoft Excel and the stress index scores were calculated for each participant using the scoresheet for each of the three specific OSI questionnaires. Categorical variables were reported as frequencies (percentages) and continuous variables were summarized as means (standard deviations). Means among the three medical workers were analysed by one-way analysis of variance (ANOVA). Between-group differences in discrete variables were assessed by using the Chi-square test. If expected cell size was under five, the Fisher's exact test was used with groups take two at a time. Tukey's HSD post hoc test was calculated for each mean comparison. A correlation matrix was used to investigate the dependence between multiple variables at the same time. All analyses were performed using the R statistical packages. A p-values <0.05 was considered statistically significant.

| Validity, reliability and rigour
The OSI is an international validated survey tool developed and tailored specifically for professionals (Belkic, 2000;Belkic & Nedic, 2007 ; Levi et al., 2000;Nedic et al., 2010) to assess precisely levels of known stressors experienced by medical workers (including nurses, doctors, IVF laboratory technicians). The English versions of OSI questionnaire were translated into Vietnamese by using double translation technique approved by original developer in the US for ensuring linguistic validity.
Cronbach's alpha was also calculated for variable elements of each aspect of the OSI and was used to measure the strength of internal consistency of OSI aspects. The Cronbach alpha of OSI aspects which ranged from 0.75 to 0.79 in our findings showed acceptable reliability.

| RE SULTS
Of the invited 131 medical professionals, 105 (80%) completed the confidential self-administered questionnaire ( Figure 1). Thirty-five participants (33.3%) were nurses, 19 (18.1%) were doctors and 51 (48.6%) were IVF laboratory technicians. Approximately two-thirds reported not having children (67.6%, 71), half (50.48%, 53) were married and three-quarters (76.2%, 80) were women, with a significant difference by medical worker group (p < 0.05). Among the three groups, nurses significantly had lower education levels and lower income compared with the others. Further demographic characteristics of all participants are given in Table 1. Some of them (e.g., income, long working hours, education level, autonomous workspace) statistically represented the high significant source of job stress.
The OSI scores were significantly different among three groups with the exception of the threat avoidant vigilance, with nurses scoring most highly (see Table 2). Figure 2 shows where the differences in mean levels of OSI aspects occurred among the three groups. In nurses, the OSI score only had a moderate one with the noxious physical exposures, the threat avoidant vigilance and the conflict (p < 0.05) and had a strong uphill linear pattern with the high demand (p < 0.001). In IVF laboratory technicians, the threat F I G U R E 1 Recruitment process flowchart avoidant vigilance and the conflict had a strong relationship with OSI score (p < 0.001) while the strictness and the noxious physical exposures had a moderate one (p < 0.001). Otherwise, the OSI score had a strong relationship with the high demand (p < 0.05) and had a moderate on with the conflict (p = 0.05) and the underload (p = 0.16) in the physician group. The detailed stressors between medical worker groups were clearly analysed in Table 3.

| D ISCUSS I ON
This is the first study to examine stressors for health workers in IVF Clinics. What we have shown is that there are clear differences between nurses, doctors and IVF technicians in terms of the factors which cause them stress and each group varies in the factors that ameliorate their stress. In the nurse group, it was found that (i) the mean OSI score was significantly higher when compared with each of the other groups (p < 0.001); (ii) the OSI score only had a strong relationship with the "high demand" (p < 0.001); (iii) Some demographic variables (e.g., income, long working hours, education level, autonomous workspace) statistically represented the high significant source of job stress.
This study confirms previous findings that night shift and long working hours are one aspect of "high demand," which has a strong influence of on job stress (Cheng, Liou, Tsai, & Chang, 2015;Coffey, Skipper, & Jung, 1988;Kirkcaldy, Trimpop, & Cooper, 1997;Nabirye, Brown, Pryor, & Maples, 2011;van Wijk, 1997). Our results are also consistent with the findings of Hsiu-Yueh when the occupational stress differed significantly by salary level (Hsu et al., 2010).
However, the findings of Garrett and McDaniel showed that demographic variables (e.g., education level, years of service) had no statistically significant relationship with job stress (Garrett & McDaniel, 2001).
As mentioned above, some demographic variables statistically represented the high significant source of job stress. We did a further linear regression analysis to check whether the significant difference between groups is caused by these demographics. The findings show that demographic variables were not the factors causing these TA B L E 1 Baseline characteristics of the three medical worker groups

| Limitations
Although the OSI is an international detailed and validated survey tool developed specifically for professionals, this study also has several limitations. First, the recall and nonresponse bias maybe occurred because the questionnaire measured only self-reported data and was quite long with nine pages. Second, the study was only based on quantitative approach. Future studies should include a qualitative aspect to gain a deep understanding of these issues.
Third, this study failed to differentiate normal nurses from the stress ones. The Vietnamese validated version of OSI questionnaire is needed to find the cut-off score for appropriate and just-in-time intervention.

| CON CLUS IONS
In the nurse group, the OSI score was significantly highest among the three groups. This points to the importance of "high demand" aspect which has a strong relationship with OSI score. The authors recommend that further studies involve not only nurses but also medical workers in IVF field to increase the generalizability of the findings.

ACK N OWLED G EM ENTS
The following are acknowledged for Karen Belkic (Karolinska Institute, Claremont Graduate University, University of Southern California) who so generously assisted us in this study (e.g., double translated OSI questionnaires into Vietnamese), and research staff of My Duc Hospital who assisted with data collection.

CO N FLI C T O F I NTE R E S T S
No conflict of interest has been declared by the authors. • substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; • drafting the article or revising it critically for important intellectual content.