Do multiple personal roles promote working energetically in female nurses? A cross-sectional study of relevant factors promoting work engagement in female nurses

We aimed to clarify the current state of work engagement in female nurses and its relationship with factors, such as the presence of multiple roles.

Subjects were 1225 female nurses working at three general hospitals, each with at least 200 hospital beds in Fukuoka Prefecture, Japan. The study was a cross-sectional design using anonymous self-administered questionnaires. A total of 1225 questionnaires were distributed and 650 responses were received (response rate 53.1%). Of those, 612 were valid (valid response rate 50.0%). The 612 subjects who provided valid responses were divided into two groups based on the presence or absence of multiple roles. Multiple roles in this study were defined as work roles and roles played as a result of life events such as marriage, childbirth, and child-rearing. The main roles refer to worker roles and wife or mother roles, as life events such as marriage, childbirth, and child-rearing in Japan can lead to women discontinuing their nursing profession. As a result, 262 nurses had multiple roles and 350 did not. We calculated the power based on previous studies and confirmed that the minimum sample size needs to be at least 100. In this study, the number of female nurses with multiple roles was 262, indicating an appropriate sample size.

While distributing and collecting the questionnaires, subjects were provided with a written explanation about the purpose and methods of the study and were assured that participation was voluntary. Submission of a completed questionnaire was considered to indicate consent. Questionnaires were returned via post.

The entire procedure followed in the study was reviewed and approved by the Ethics Committee of the University of Occupational Environmental and Health Japan (Approval No. H28-156).

Work engagement was assessed using the short-form Japanese version of the Utrecht Work Engagement Scale (UWES-J) created by Shimazu et al. based on the simplified version of the original UWES created by Schaufeli and Bakker [16, 34]. The short-form UWES-J has three subscales—“vigor,” “dedication,” and “absorption”—and comprises nine items on a 7-point scale. Higher scores correspond to stronger work engagement. The reliability and validity of the UWES-J have been demonstrated in previous studies (Cronbach’s alpha = 0.92) [35‐37]; in the current study, the Cronbach’s alpha was 0.94.

Job demands and job resources were assessed with items from the new Brief Job Stress Questionnaire [38]. The job demands section comprised 13 items on topics such as quantitative and qualitative workload and role conflict. The job resources section comprised 27 items on topics such as job resources and support from bosses and colleagues. Items in both sections were assessed on a 4-point scale with possible responses of “strongly agree,” “agree,” “disagree,” and “strongly disagree.” Higher scores corresponded to more job demands and more job resources, respectively. The reliability and validity of the new Brief Job Stress Questionnaire have been demonstrated in a previous study (Cronbach’s alpha > 0.70) [39]; in the current study, the Cronbach’s alpha was 0.81.

Spillover effects were measured using the Japanese version of the Survey Work–Home Interaction-NijmeGen (SWING) created by Geurts et al. [40] and translated by Shimada, Shimazu, Kawakami, et al. [41] The Japanese version of the SWING (SWING-J) has four subscales (WFNS, FWNS, WFPS, and FWPS) comprising 22 items measured on a 4-point scale with possible responses of “never,” “sometimes,” “often,” and “always.” Higher scores correspond to more spillover in each scale, respectively. This scale measures the balance between “work” and “home,” which includes spouse, family members, and friends. The reliability and validity of the SWING-J have been demonstrated in a previous study (Cronbach’s alpha = 0.75–0.86) [41]; in the current study, the Cronbach’s alpha was 0.83.

Characteristics of coping, which is a personal resource, were assessed with the Brief Scales for Coping Profile (BSCP) for workers created by Kageyama et al. [42‐44] The BSCP has six subscales: “active solution for problems” and “seeking help for solution” as problem-focused coping and “changing mood,” “changing a point of view,” “emotional expression involving others,” and “avoidance and suppression” as emotion-focused coping. It comprises 18 questions measured on a 4-point scale with possible responses of “rarely,” “occasionally,” “sometimes,” and “often.” Higher scores corresponded to more frequent use of that type of coping. The reliability and validity of BSCP have been demonstrated in previous studies (Cronbach’s alpha = 0.66–0.79) [43–45]; in the current study, the Cronbach’s alpha was 0.77.

Attributes investigated were age, years of experience, frequency of night shifts per month, position, type of employment, department, shift format, marital status, presence or absence of a spouse, cohabitation status with spouse, employment of spouse, presence or absence of children, number of children, presence or absence of children under school age, and the number of children under school age.

Descriptive statistics were calculated for attributes and scores for each scale. To compare the two groups (presence and absence of multiple roles), Fisher’s exact tests and chi-square tests were performed on the subjects with each attribute, and Mann–Whitney U tests on the scores of each scale. For all subjects, simple correlation tests were performed with work engagement as the dependent variable and each measure as an independent variable. Next, hierarchical multiple regression analysis was performed on variables found to be significantly associated with work engagement in simple correlation tests (p < 0.05) and variables found to differ between groups (p < 0.05) with years of experience and presence or absence of multiple roles as independent variables. The Kolmogorov–Smirnov test was used to assess the normality of work engagement as a dependent variable. If the data were not normally distributed, the distribution was corrected with Blom’s normal score transformation. Variables were included by forced entry for each category, and a dummy variable was included for the presence or absence of multiple roles (0, without multiple roles; 1, with multiple roles). Multicollinearity between variables was verified with a criterion of variance inflation factor < 2. The analysis was performed using SPSS Statistics version 24 statistical software (IBM).

Based on a previous study, the reference score for the UWES short form was 3.7 points [48], and the mean score of subjects in the present study was 2.6 points. Further, this is roughly the same as the mean score of 2.6 points for women in Japan [6] and 2.6 to 2.8 points for nurses according to previous studies [49‐51]. Therefore, it was concluded that the work engagement of subjects in the present study was similarly low as that of female workers and nurses in Japan in general.

The present study showed higher work engagement among female nurses with multiple roles than those without. The female nurses with multiple roles in this study had experienced life events such as marriage, childbirth, and child-rearing, and held roles such as a wife and mother. Previous studies have also found that work engagement is higher among (a) female nurses who are married than those who are not married [25] (b) married female nurses with four or more years’ experience [5], and that nurses with children have a higher sense of achievement regarding work and are less likely to experience burnout, the polar opposite of work engagement [26]. The results of the present study support those of previous studies. Moreover, female nurses with multiple roles were older and more experienced than those without. Previous studies have also found that work engagement increases with increasing age and experience [6, 50] suggesting that the difference in work engagement between the two groups may be influenced by age and years of experience.

In the hierarchical multiple regression analyses, work engagement of female nurses, which was the dependent variable, was predicted by high FWPS, more job resources, greater use of “changing a point of view” and “active solution for problems” as coping strategies, the presence of multiple roles, and less use of “avoidance and suppression” as a coping strategy, even after controlling for years of experience.

In the present study, based on the JD-R model, we postulated that spillover effects that include variables in both the work and family domains would be associated with higher work engagement, which is an outcome in terms of high WFPS and FWPS, and low WFNS and FWNS. However, for positive spillover, only FWPS, namely, the positive effects of family on work, was associated with work engagement, and there were no associations with any type of negative spillover. A comparison of the spillover effects between the two groups revealed that FWPS was higher among those with multiple roles than those without. This suggests that female nurses with multiple roles may be more influenced by the effects of FWPS, which is a resource that increases work engagement, than by the presence of multiple roles. The circumstances within the family domain that positively affect the work domain may be an important contributing factor. A study on the determinants of work engagement in nurses working in hospitals found that the presence of someone who is pleased by their work contributes to higher work engagement [52]. Female nurses with multiple roles, having family members with whom they share their daily lives and who acknowledge and delight in their work, may build better relationships and increase FWPS, leading to a further increase in work engagement.

Coping strategies that were associated with an increase in work engagement included the greater use of “changing a point of view” and “active solution for problems” strategies, and reduced use of “avoidance and suppression.” These results are consistent with earlier work studying factors related to work engagement of nurses working at hospitals examined by years of experience; this research suggested a greater use of “changing a point of view” and less use of “avoidance and suppression” strategies among nurses with between 4 and 9 years’ experience and 10 or more years’ experience [5]. Another study showed that greater use of problem-focused coping strategies and less use of avoidance coping strategies was related to better work engagement in nurses [53]. Further, according to cross-sectional data on workers, those who often used problem-focused coping strategies or “changing a point of view” as a strategy had stronger stress-mediating factors in the workplace, such as level of control, level of achievement, and support from coworkers. On the other hand, workers who often used an avoidance coping strategy felt more stress from interpersonal relationships and a higher qualitative workload [42]. Moreover, daily rumination has been found to inhibit the use of contextual resources [54]. The current results suggest that having more job resources is related to better work engagement in female nurses, consistent with the JD-R model [3, 4], but that simply having more job resources may not be enough; those resources may not be utilized depending on the types of coping strategies used by the worker. While it is necessary to confront issues and actively seek resolutions to problems, an attitude of avoiding ruminating on one problem for too long and changing perspectives may lead to greater use of job resources, and thus increase work engagement.

Compared to the scores of the standard population calculated in the BSCP development process (9.6 points for “active solution for problems,” 8.0 points for “seeking help for solution,” 7.6 points for “changing mood,” 7.7 points for “changing a point of view,” 4.4 points for “emotional expression involving others,” and 6.4 points for “avoidance and suppression”) [44], the present study showed higher scores among all subjects in “seeking help for solution,” “changing mood,” and “emotional expression involving others,” and lower scores in “active solution for problems,” “changing a point of view,” and “avoidance and suppression,” all of which are related to an increase in work engagement. This suggests that the greater use of “active solution for problems” and “changing a point of view” and the reduced use of “avoidance and suppression” may be important coping strategies as personal resources to enable female nurses to work energetically.

Based on the difference in the presence or absence of multiple roles, nurses with multiple roles had lower scores than the standard population in “changing a point of view” and “avoidance and suppression” and roughly the same scores in “active solution for problems” and “changing mood.” In contrast, scores for “active solution for problems,” “changing a point of view,” and “avoidance and suppression” were lower among nurses without multiple roles compared to the standard population. This indicates that the greater use of “active solution for problems” and “changing a point of view” as coping strategies may increase work engagement in female nurses. In addition, while it has been suggested that the use of both “active solution for problems” and “changing mood” are effective in improving mental health associated with work engagement [55], nurses with multiple roles did not use “changing mood” more often than nurses without multiple roles. This may be explained by the fact that nurses with multiple roles have a spouse and possibly children, unlike nurses without multiple roles, and may be less able to spend time or money on themselves, thus making it difficult to find the time to “change their mood.” [56] Furthermore, nurses with multiple roles did not use “avoidance and suppression” more as compared to those without multiple roles or even compared with the standard population. This may be because nurses with multiple roles cannot avoid responding to the demands of their family and have no choice but to address them in their day-to-day lives. Therefore, it is speculated that one is unable to avoid stressors when coping with them, leading to the realization from daily life experiences that avoiding stressors is not an effective strategy to cope with them. This may lead to a greater tendency in nurses with multiple roles to feel that they have many job resources, and a reduced tendency to feel that there are heavy job demands.

In a similar vein, having greater “job resources” was detected as a determinant of work engagement in the work domain, and this is consistent with the JD-R model [3, 4]. Like previous studies, the current findings suggest that various resources in the work domain may increase positive feelings towards work in female nurses. However, the score for “job resources” of the subjects was lower than that of female employees working in the healthcare and welfare industry in Japan [57]. “Job resources” in the present study refers to resources at the work operation level, department level, and workplace level. Therefore, helping nurses in Japan work energetically may require improving nursing care work as well as the organizational work environment (e.g., for the hospital as a whole and the nurse’s department).

This study has some limitations. First, the participants in this study were limited to those from a single prefecture in Japan, and the response rate was about 50%. Thus, the sample may not be representative of all female nurses in Japan. Although age, years of experience, and employment conditions of the participants in this study are very similar to the average in Japan, further studies with a higher response rate are needed, and collecting more comprehensive data could provide insights into how female nurses can lead a more balanced and active work life.

Second, although we used the results of model 5 that had the largest adjusted R² value, it was still only 0.324; therefore, we were able to explain work engagement with predicted variables by only about 30%. The determinants of work engagement used in the present study were presence or absence of multiple roles, coping strategies as a variable related to personal resources, job resources and job demands as variables in the work domain, and spillover effects as a variable in the work and family domain. However, other relevant variables may also influence the level of work engagement [58]. Further research should be conducted to investigate additional variables that may help determine work engagement. Above all, we collected data concerning participants’ evaluation of salaries, which were included in the job resources domain. However, in the future, additional information should be analyzed, including the salaries of nurses’ spouses, household income, and nurses’ satisfaction with both.

Third, the presence or absence of multiple roles was significantly associated with work engagement, but the standardized regression coefficient β was as low as 0.097. This is because multiple roles in this study only include employee, wife, and mother, and caregiver role was not included. Therefore, studies that include caregiver role as part of multiple roles should be conducted in the future.

Lastly, the survey was conducted only with women. Most nurses in Japan are women; however, expanding the scope of research to include men would further elucidate nurses’ work engagement and could help determine the measures that are necessary to improve the quality of nursing care.