Introduction
Nurses in health-care environments, in daily contact with suffering patients, terminal situations, and impacting treatments, are one of the categories most exposed to the risk of work stress, turnover, dissatisfaction, and burnout (BO) (e.g., Gama et al. Reference Gama, Barbosa and Vieira2014; Gómez-Urquiza et al. Reference Gómez-Urquiza, Albendín-García and Velando-Soriano2020; Jennings Reference Jennings and Hughes2008); the main organizational determinants of negative work outcomes among healthcare workers (HCWs) include the quality of working conditions, interpersonal relationships, role conflict, and high work demands (e.g., Gómez-Urquiza et al. Reference Gómez-Urquiza, Albendín-García and Velando-Soriano2020; Maslach et al. Reference Maslach, Schaufeli and Michael2001; Rizo-Baeza et al. Reference Rizo-Baeza, Mendiola-Infante and Sepehri2018); among the organizational factors that are instead considered protective of negative outcomes, employee perception of how attentive the organization is in evaluating and enhancing both the contributions received from its workers and their state of well-being (perceived organizational support [POS]) appears to have a relevant role (Bao and Zhong Reference Bao and Zhong2019; Xu and Yang Reference Xu and Yang2021; Yi et al. Reference Yi, Kim and Akter2018; Zeng et al. Reference Zeng, Zhang and Chen2020).
Numerous studies have explored the role of cognitive and emotional demands and resources on workers’ outcomes (i.e., emotional exhaustion, commitment, job satisfaction, etc.) within the job demands-resources (JD-R) model (e.g., Bakker and Demerouti Reference Bakker and Demerouti2017; Li et al. Reference Li, Tuckey and Bakker2022).
The model (Schaufeli and Taris Reference Schaufeli, Taris, Bauer and Hämmig2014; Xanthopoulou et al. Reference Xanthopoulou, Bakker and Demerouti2012) hypothesizes the development of job strain and BO when the individual perceives an imbalance between JD and resources at work (Bakker and Demerouti Reference Bakker and Demerouti2017); the JD-R model, furthermore, theorizes that work outcomes are the result of the interaction between work demands and resources (i.e., support) to deal with them.
Based on the JD-R model, correlational research was designed with the participation of a homogeneous sample of oncology nurses from an Italian public hospital, who filled out a questionnaire that investigated the relationships between JD, POS, intended as job resource, self-esteem, as personal resource, and job outcomes (BO and job satisfaction). More specifically, as regards the theoretical contribution, the research aimed to test the JD-R model on the sample of Italian oncology workers, and the role of POS as a moderator of the effects of JD on outcomes.
The results of the research can provide indications both for the development of practices for managerial and peer support, the implementation of policies for the prevention of BO (e.g., training, empowerment, prevention, team building, operational solutions, etc.), and one-to-one support tools (e.g., sick days, absences, turnover, etc.) (Ahmad et al. Reference Ahmad, Barattucci and Ramayah2022; Crawford et al. Reference Crawford, LePine and Rich2010; Serban et al. Reference Serban, Rubenstein and Bosco2022).
JD-R model in the oncology setting
The JD-R model (Bakker and Demerouti Reference Bakker and Demerouti2017; Brauchli et al. Reference Brauchli, Jenny and Füllemann2015) is one of the most used conceptual frameworks in the study of the relationship between organizational factors, personal factors, and job outcomes. According to this model, the work environment is composed of: (a) JD (physical, social, and organizational factors that require effort and costs) (Bakker and Demerouti Reference Bakker and Demerouti2017; Bakker and de Vries Reference Bakker and de Vries2021) which are not necessarily negative but may require an activation effort; (b) job resources which represent a set of variables of different nature (organizational, relational, and psychological) that have a positive relationship with work outcomes, but an inverse relationship with work demands (Bakker et al. Reference Bakker, Hakanen and Demerouti2007; Broetje et al. Reference Broetje, Jenny and Bauer2020).
There are a limited number of studies that have used the JD-R model on samples of oncology HCWs, despite this being one of the categories with the highest levels of JD (Costeira et al. Reference Costeira, Ventura and Pais2022; Wazqar Reference Wazqar2018), BO, and dissatisfaction (Adil and Baig Reference Adil and Baig2018; Lazarescu et al. Reference Lazarescu, Dubray and Joulakian2018).
Job resources and the role of perceived organization support
Recent studies underline that POS, as an organizational resource (Kurtessis et al. Reference Kurtessis, Eisenberger and Ford2017; Lee and Peccei Reference Lee and Peccei2007), may impact work outcomes, and play a role in perceptions related to JD.
The POS represents the set of perceptions of the worker relating to how attentive the organization is to the aspects of well-being, operational support, and staff development (Zeng et al. Reference Zeng, Zhang and Chen2020). Organizational support theory (Caesens and Stinglhamber Reference Caesens and Stinglhamber2020; Rhoades and Eisenberger Reference Rhoades and Eisenberger2002) theorizes that perceptions of support are determined not only by aspects of operational management support, but also by environmental, remuneration, and fairness aspects.
In HCWs, a high level of POS seems generally associated with better job outcomes and a positive psychological state (e.g., Zeng et al. Reference Zeng, Zhang and Chen2020), lower strain symptoms such as anxiety, fatigue, and BO (Grama and Băiaș Reference Grama and Băiaș2018; Lecca et al. Reference Lecca, Finstad and Traversini2020; Rhoades and Eisenberger Reference Rhoades and Eisenberger2002; Wu et al. Reference Wu, Singh-Carlson and Odell2016), and higher job satisfaction (Canboy et al. Reference Canboy, Tillou and Barzantny2021; Kurtessis et al. Reference Kurtessis, Eisenberger and Ford2017). Similar indications have been provided by other research on nurses in public and private hospitals (Özyer et al. Reference Özyer, Berk and Polatcı2016; Riggle et al. Reference Riggle, Edmondson and Hansenc2009); however, few studies have explored the relationship between POS and job outcomes in oncology workers (Guveli et al. Reference Guveli, Anuk and Oflaz2015; Head et al. Reference Head, Middleton and Zeigler2019; Yi et al. Reference Yi, Kim and Akter2018).
Referring to recent evolutions of the JD-R model, furthermore, some job resources seem to have a possible moderating role on the effects of JD on outcomes (e.g., Bao and Zhong Reference Bao and Zhong2019; Tummers and Bakker Reference Tummers and Bakker2021; Xu and Yang Reference Xu and Yang2021); however, very few contributions on the health-care sector have tested it, and equally few with a sample of nurses; since several studies in other work sectors have confirmed the role of POS as a possible moderator between some organizational determinants of stress and outcomes, it appears useful to explore these relationships in samples of oncology nurses (Canboy et al. Reference Canboy, Tillou and Barzantny2021; Serban et al. Reference Serban, Rubenstein and Bosco2022).
Personal resources: The role of self-esteem
Among the so-called personal resources, self-esteem has developed some interest in the JD-R model of work outcomes; HCWs with higher self-esteem seem to better cope with job stress, as having a healthy confidence in one’s skills and self-concept helps them to put in place efficient stress management with clarity and composure of thought. Since it is clear that self-esteem affects the way people consider themselves, and influences their professional development, some studies have investigated its relationship with work outcomes (Johnson et al. Reference Johnson, Jayappa and James2020; Kupcewicz and Jóźwik Reference Kupcewicz and Jóźwik2020; Molero Jurado et al. Reference Molero Jurado, Pérez-Fuentes and Gázquez Linares2018). In HCWs,self-esteem is proven to be positively associated with job satisfaction (Lee and Peccei Reference Lee and Peccei2007), as well as for intensive care unit nurses (Liu et al. Reference Liu, Zhang and Chang2017); nevertheless, no research has yet examined the role of self-esteem on the outcomes of oncology nurses. Based on the above-described framework and rationale (Bakker and Demerouti Reference Bakker and Demerouti2017), the present research integrated insights provided by literature (Serban et al. Reference Serban, Rubenstein and Bosco2022; Turnell et al. Reference Turnell, Rasmussen and Butow2016; Zeng et al. Reference Zeng, Zhang and Chen2020) in a model in which JD, POS (as an organizational resource), and self-esteem (as a personal resource) are considered as antecedents, and BO and job satisfaction as the outcomes (Figure 1).
Figure 1. Tested conceptual model (the colored lines refer to moderated effects): moderation model in which the effect of both determinants (JD and POS) on outcomes is moderated by the other determinant.
Given the assumptions of the JD-R model (Bakker and Demerouti Reference Bakker and Demerouti2017), and based on the evidence provided in the literature, we, therefore, hypothesized as follows:
Hp1: Job demands will significantly predict job outcomes; more specifically, we expect that high levels of JD will positively predict high BO levels (Hp1a), and negatively high job satisfaction levels (Hp1b).
Overall, from a review of the literature, it is clear that in HCWs, support from colleagues and from management is negatively associated with BO (Rizo-Baeza et al. Reference Rizo-Baeza, Mendiola-Infante and Sepehri2018; Wazqar Reference Wazqar2018), and positively with job satisfaction (Assiri et al. Reference Assiri, Shehata and Assiri2020; Courtnage et al. Reference Courtnage, Bates and Armstrong2020; Kitajima et al. Reference Kitajima, Miyata and Tamura2020). Consequently, it seems correct to assume that:
Hp2: POS will significantly predict job outcomes; more specifically, we hypothesize that high levels of POS will negatively predict high BO levels (Hp1a), and positively high job satisfaction levels (Hp2b).
Based on the indications provided by the theoretical framework and previous research (Costeira et al. Reference Costeira, Ventura and Pais2022; Gama et al. Reference Gama, Barbosa and Vieira2014), it is possible to assume that:
Hp3: Self-esteem will significantly predict job outcomes; more specifically, we hypothesize that high levels of self-esteem will negatively predict high BO levels (Hp3a), and positively high job satisfaction levels (Hp3b).
As a result of the abovementioned indications and again referring to the JD-R model, we hypothesized as follows:
Hp4: The interaction effect of JD × POS will be significant both for BO (Hp4a) and job satisfaction (Hp4b). The association between JD and outcomes will vary as a function of POS levels. The tested model and research assumptions are shown in Figure 1.
Methods
Research design
This research adopts a quantitative approach with a cross-sectional design.
Participants
The Italian National Health Service guarantees health care for cancer patients, the provision of palliative care, and through collaboration with the rich network of voluntary associations, also guarantees home care. The hospital, with over 1500 employees, operates in an area with a high population density. The oncology department is organized into 3 services: an oncology hospital, an operational unit, and a hospice.
Measures
Participants completed the first section of the questionnaire with sociodemographic information and then they filled out a questionnaire made up of the following measures:
The POS scale, originally developed by Eisenberger et al. (Reference Eisenberger, Wang and Mesdaghinia2014), and based on literature indications (the majority of studies on POS use a short form developed from the 17 highest-loading items in the POS; Eisenberger et al. Reference Eisenberger, Shanock and Wen2020), is a scale that measures the perceptions of beneficial treatment received by employees (e.g., “The organization where I work cares about my mental and physical well-being”). In the present study, POS was measured by the Italian version (8 items) (Di Stefano et al. Reference Di Stefano, Venza and Aiello2020; Muse and Stamper Reference Muse and Stamper2007). All items were on a 6-point scale ranging from 1 (strongly disagree) to 6 (strongly agree). Cronbach’s alpha coefficient was .95.
The Rosenberg Self-Esteem Scale (Rosenberg Reference Rosenberg1965; Italian adaptation by Sartirana et al. Reference Sartirana, Camporese and Dalle Grave2013), is a 10-item scale that estimates global self-worth by measuring positive and negative feelings about the self (e.g., “I feel that I have a number of good qualities”). Items are answered using a 4-point Likert scale format ranging from “strongly agree” to “strongly disagree.” Cronbach’s alpha coefficient was .79.
The Professional Quality of Life Scale (ProQoL; Stamm Reference Stamm2009; Italian adaptation by Palestini et al. Reference Palestini, Prati and Pietrantoni2009), aims to gauge the professional quality of life, through the measurement of 3 aspects of professional quality of life: compassion satisfaction (CS), compassion fatigue, and BO. In the present study, only the CS and BO dimensions of the ProQOL were utilized. The satisfaction subscale (8 items) measures the employees’ satisfaction with their ability (e.g., “My work makes me feel satisfied”). The BO subscale (7 items) measures if the worker is experiencing symptoms of BO (e.g., “I feel worn out because of my work”). Items were rated on a 5-point scale. Participants were asked: in the last month how many times, ranging from 1 (never) to 5 (very often). Cronbach’s alpha coefficient for CS was .86, and .88 for BO.
JD were measured with a scale from the literature (Bakker and Demerouti Reference Bakker and Demerouti2017; Lesener et al. Reference Lesener, Gusy and Wolter2019), that measures work pressure and emotional demands. (e.g., “I have to work very fast/my job requires me to keep a lot of information in mind at once”). In the present study, JD was measured by the Job Demands Italian 27-item scale version (De Carlo et al. Reference De Carlo, Falco and Capozza2008). The items were rated on a 6-point scale ranging from 1 (strongly disagree) to 6 (strongly agree.) The scale is believed to be unidimensional. Cronbach’s alpha coefficient was .83.
Sociodemographic variables, participants were asked to give information on their sociodemographic characteristics, such as gender, age, education, marital status, shift work, and seniority.
In order to address response bias and common method variance, we recurred to the suggested methods in literature (Baumgartner et al. Reference Baumgartner, Weijters and Pieters2021; Kock et al. Reference Kock, Berbekova and Assaf2021; Podsakoff et al. Reference Podsakoff, MacKenzie and Lee2003) and various scale endpoints and formats for the measured variables were used to reduce method biases caused by commonalities in scale endpoints and anchoring effects, and scales were graphically separated.
Data analyses
The analytical approach was correlational. Cronbach’s alphas and zero-order correlations were used to assess the scales’ internal consistencies and examine associations between pairs of continuous variables; with the purpose of exploring the differences in the measured variables related to sociodemographic and work variables, independent sample t-tests, Analysis of Variance (ANOVAs), and correlational analysis were carried out, using IBM SPSS 23. Relationships between measured variables were examined through correlation analysis and multiple regressions, using SPSS 23 and SPSS PROCESS Macro 3.3.
More specifically, 2 moderation analyses were run to verify whether POS moderated the relationship between JD (and vice versa) and outcomes (BO and job satisfaction). For each analysis, PROCESS model number 1 with the macro developed by Hayes was run, estimating the relationship between the predictor and the criterion at low, medium, and high levels; the PROCESS macro allows bootstrapping (Hayes Reference Hayes2018), a nonparametric resampling procedure that does not assume normality and involves the extraction of several thousand subsamples (5000, in the present case) from a dataset. Through bootstrapping, the distribution of effects is empirically approximated and used for calculating confidence intervals (Preacher and Hayes Reference Preacher and Hayes2004). For each association, the unstandardized B coefficient along with the 90% lower and upper limits of its respective confidence interval will be provided. Interactions were probed through the Johnson–Neyman technique. This technique provides a region of significance of the effect of X on Y; that is, it provides a continuum where the conditional effect of X on Y transitions between statistically significant and not significant at the alpha level of significance (Hayes Reference Hayes2018).
Results
Recruitment
The hospital was approached through a formal request to participate in the project, which was presented to managers and the head of the unit. A project on “Work-related stress and organizational resources” dedicated to all the nurses (N = 275) in the oncology ward of a Sicilian public hospital started in June 2021; a cross-sectional study was then carried out from 10 September 2022 to the end of December 2023, involving 262 voluntarily participating nurses from 3 different units: oncology hospital, hospice, and operational unit, from the same geographical area (Southern Italy) (Table 1). At the end of a short training meeting conducted by 2 researchers relating to the aforementioned project, during working hours, the nurses were given a paper and pencil questionnaire to complete and return in 5 days. A link was sent to complete the questionnaire online via Google form to those absent from the meeting. Missing data treatment was necessary (questionnaires completed with a missing percentage greater than 5%) and reduced the final analysis sample from 262 to 235 nurses.
Table 1. Sample description
Descriptives
The final analysis sample was made up of 235 nurses working in the oncology ward of a Sicilian public hospital, mostly in the hospice (N = 145, 62.7%), who completed the questionnaire in full. The sample was fairly balanced by gender (N = 124 women, 52.4%), with an average age of 46.47 years (SD = 8.36), mostly married (N = 164, 69.9%), with children (N = 176, 74.9%), and graduates (N = 127, 54.2%). Nurses mainly worked shifts (N = 209, 89.3%), including night shifts (N = 158, 67.5%), and the average seniority was 14.1 years (SD = 9.2).
No gender differences occurred for any of the variables, and no relationships between age or seniority and measured variables resulted from statistical analyses. Moreover, the ANOVA did not reveal any significant differences between groups regarding the level of education, marital status, and work structure for any of the considered variables. A barely significant difference between nurses with night and day shifts was found for JD (t 233 = −2.1; p < .05; day shift, mean = 4.02, SD = .98; night shift, mean = 4.39, SD = .87).
Correlational analysis
JD resulted significantly positively correlated with BO and negatively with job satisfaction; moreover, both POS and self-esteem were significantly negatively correlated with BO and positively with job satisfaction. Table 2 depicts descriptive statistics and correlations between study variables.
Table 2. Descriptive statistics (mean and standard deviation) and correlations between measured variables
Note: r, Pearson correlation coefficient;
* p < .05, **p < .01, ***p < .001.
Regression analysis
With the aim of testing hypotheses, 2 multiple linear regressions including BO and job satisfaction as criterion variables, JD, POS, and self-esteem as main predictors, were performed (Table 3).
BO was positively predicted by JD, and negatively by POS and self-esteem, confirming hypotheses 1a, 2a, and 3a. The interaction term (POS × JD) was statistically significant (Table 3), therefore, confirming hypothesis 4a. Predictors explained about 27% of the BO’s variance.
Table 3. Outcomes regressed on measured antecedents
Note:
** p < .01, ***p < .001.
Regression analysis revealed that job satisfaction was positively predicted by POS and self-esteem, and negatively by JD, thus confirming hypothesis 1b, 2b, and 3b. The interaction term (POS × JD) was statistically significant, confirming hypothesis 4b. Predictors explained about 23% of the job satisfaction’s variance.
Moderation role of POS between JD and outcomes
The interaction between JD and POS, with respect to outcomes, was probed through the Johnson–Neyman technique (value = 30.03). The positive association between JD and BO was significant at low (b = .69, CI: [.379, 1.011]) and medium levels of POS (b = .43, CI: [.246, .631], while it was not significant at high levels of POS (b = .22, CI: [−.016, .464]) (Table 4 and Figure 2); the overall equation was significant (R 2 = .16; F (3, 230) = 33.01; p < .000), and the JD by POS interaction significantly increased the explained variance (R 2 change = .027, F (1, 232) = 5.40; p = .021). This outcome indicates that the more the nurses perceive a supportive organization, the weaker the positive relationship between JD and BO (Figure 2).
Table 4. Result of regression analysis concerning the moderation effect of POS on the job demands-burnout relationship and the conditional influence of POS based on the Johnson–Neyman technique
Figure 2. The association between job demands and burnout as a function of POS.
The negative association between JD and satisfaction was significant at low (b = .49, CI: [.308, .688]) and medium levels of POS (b = .29, CI: [.179, .41], while it was not significant at high levels of POS (b = .125, CI: [−.018, .269]) (Table 5 and Figure 3); the overall equation was significant (R 2 = .18; F (3, 230) = 12.42; p < .000), and the JD by POS interaction significantly increased the explained variance (R 2 change = .047, F (1, 232) = 9.37; p = .0026). This outcome indicates that the more the nurses perceive a supportive organization, the weaker the negative relationship between JD and satisfaction (Figure 3).
Figure 3. The association between job demands and satisfaction as a function of POS.
Table 5. Result of regression analysis concerning the moderation effect of POS on the job demands-satisfaction relationship and the conditional influence of POS based on the Johnson–Neyman technique
Discussion
Oncology nurses are particularly prone to BO and dissatisfaction due to various professional practice circumstances and working conditions that can lead to physical and emotional exhaustion, turnover, or sick leave (e.g., Woo et al. Reference Woo, Ho and Tang2020): they are asked to provide care with patience and empathy (Khamisa et al. Reference Khamisa, Oldenburg and Peltzer2015) and at the same time, must cope with a stressful environment as a result of daily contact with patient suffering and emotional demands.
In line with the theoretical framework, results confirmed that JD predict both BO and job satisfaction, in the present sample of oncology nurses; POS, on the other hand, is a weak predictor of job outcomes, but the results confirm the possible mediator role proposed by the JD-R model and the references in the literature (e.g., Serban et al. Reference Serban, Rubenstein and Bosco2022); self-esteem, furthermore, proves to be a predictor of BO, in particular (Johnson et al. Reference Johnson, Jayappa and James2020). Overall, the most significant result appears to be the interaction between the JD and POS: this latter may be responsible for a possible buffer effect on the relationship between JD and outcomes.
Theoretical implications
The present results are in line with the JD-R theory (Bakker et al. Reference Bakker, Xanthopoulou and Demerouti2023; Li et al. Reference Li, Tuckey and Bakker2022), confirming that outcomes are predicted by JD and consistently with the idea that high demands at work are related to poorer outcomes (e.g., satisfaction, BO) (e.g., Li et al. Reference Li, Tuckey and Bakker2022), and performance (Rao and Krishna Reference Rao and Krishna2021).
Always in line with the JD-R Model, as regards self-esteem, findings suggest that personal resources predict BO and, to a lesser extent, satisfaction levels. From an empirical standpoint, our findings are also consistent with other contributions that have shown that organizational job resources may attenuate the adverse effect of JD on positive and negative outcomes: POS may play a central role in employee’s well-being and health, acting as a possible moderator, and somehow defusing the positive association between JD and outcomes, also in oncology setting; POS is considered an organizational resource that can generate a range of positive emotional perceptions and experiences in the workplace (Özyer et al. Reference Özyer, Berk and Polatcı2016), and can replenish resources consumed by emotional labor and counter time pressure (Riggle et al. Reference Riggle, Edmondson and Hansenc2009); as also reported by other studies on oncology workers, human resource (HR) management focused on support and team collaboration will lead to workers perceiving high level of job satisfaction and lower level of BO (Courtnage et al. Reference Courtnage, Bates and Armstrong2020).
In work contexts with high emotional demands, POS appears to be able to modulate the effects of JD on outcomes with a sort of buffer effect; according to the organizational support theory (e.g., Eisenberger et al. Reference Eisenberger, Shanock and Wen2020; Xu and Yang Reference Xu and Yang2021), POS has been shown to have significant benefits for workers and organizations: high POS workers suffer less BO at work, are more inclined to return to work after injury, and show better performance indicators (Kurtessis et al. Reference Kurtessis, Eisenberger and Ford2017; Rhoades and Eisenberger Reference Rhoades and Eisenberger2002), possibly because employees value POS partly because it meets their needs for approval, esteem, and affiliation, and provides comfort during times of stress (Lecca et al. Reference Lecca, Finstad and Traversini2020). Overall, the present research provides a valuable contribution to the literature on the relationship between the main organizational and personal factors considered in the JD-R model, on positive and negative outcomes among oncology nurses.
Practical implications
The results provide indications to HR managers in oncology departments and institutions. It seems clear that it is possible to intervene on organizational and personal variables to weaken the natural impact of JD on outcomes in oncology settings. Measures should focus on primary and secondary prevention and be aimed at avoiding negative consequences for nurses and their patient’s quality care, such as job BO, as well as reduced nurse satisfaction (Riggle et al. Reference Riggle, Edmondson and Hansenc2009). Furthermore, both training and individual and organizational interventions (e.g., job design, empowerment, increasing job control, etc.), in addition to BO prevention, should focus on the optimization of the balance between JD and resources.
The need to ensure oncology workers’ well-being should involve the periodic monitoring of specific psychosocial and organizational factors linked to outcomes and motivation. Flexible training designed to generate a high level of work engagement and self-esteem (e.g., emotional strength, coping strategies, acceptance, etc.), by virtue of the feedback effect of these outcomes on organizational perceptions (e.g., perceptions of management support) and JD (Crawford et al. Reference Crawford, LePine and Rich2010; Serban et al. Reference Serban, Rubenstein and Bosco2022), should be implemented by health-care institutions. Moreover, since the type of behavior triggered by resources would lead to advantages both for the individual and the organization (Schaufeli and Taris Reference Schaufeli, Taris, Bauer and Hämmig2014), measures should focus on the exploration of emotional demands, enhancement of management-supporting activities, and personal resources (e.g., self-esteem). In organizations characterized by supportive management and sustainable HR management, workers have higher levels of job satisfaction, sense of citizenship, and loyalty, and are more inclined to share corporate values and goals; the POS as evidence that the organization intends to assist everyone’s work, but also a tool to take care of performance (Eisenberger et al. Reference Eisenberger, Malone and Presson2016).
Limitations and further research
It is important to underline that the results of this study are to be considered with caution and at the same time it is necessary to consider its various limitations.
First, the cross-sectional design of the study precludes conclusions about the possible causal direction of the observed relationships between variables. The nature (of convenience), the extension, and the homogeneity of the sample, moreover, limit the generalizability of the results, which should certainly be confirmed in similar samples in other cultural and organizational contexts. Considering some indications in the literature (Johnson et al. Reference Johnson, Jayappa and James2020; Kupcewicz and Jóźwik Reference Kupcewicz and Jóźwik2020), we preferred not to explore the possible role of moderator of self-esteem between JD and outcomes; however, given that self-esteem is also clearly related to perceptions of the work environment and good relationships at work, it is possible that the research has missed an opportunity to better explore its role in the reference model.
It is therefore necessary that future studies should: (a) given that past research suggested that these constructs may influence each other over time (e.g., Bakker et al. Reference Bakker, Xanthopoulou and Demerouti2023; Xu and Yang Reference Xu and Yang2021), investigate with longitudinal studies the relationships and interactions between JD, resources, and outcomes; (b) consider the differential role of specific dimensions of JD (e.g., cognitive, emotional, physical, etc.), different job resources (e.g., autonomy, leadership, role ambiguity, leader–member exchange, etc.), and personal resources (e.g., self-efficacy, optimism, resilience, flexibility) that may be relevant for oncology workers; (c) although the indications in the literature seem to be in line with our assumptions (e.g., Li et al. Reference Li, Tuckey and Bakker2022), explore the tested causal direction through studies with larger samples, experimental or longitudinal designs, and in different geographical and cultural contexts; (d) to overcome the limitations imposed by self-report measures, consider implementing third-party evaluations by supervisors or colleagues as well as objective data and possibly measurements of variables at different level (individual, group, team, organizational, etc.).
Conclusions
Oncology nurses are a population of workers exposed to multiple risk factors for psychological health, both environmental, relational, and role-related. The results of the present study support the need for organizations to create work environments that through favorable relationships and support at work can improve dedication to organizational objectives, prevent negative outcomes, and increase performance (Baran et al. Reference Baran, Shanock and Miller2012; Barattucci et al. Reference Barattucci, Lo Presti and Bufalino2020).
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1478951524000890.
Author contributions
TR and MB worked on the original idea and carried out the detailed conceptualization and investigation of this research. MB finalized the methodology. KC carried out the data collection. MB carried out data analysis and wrote the results section. TR, GS, and MB carried out the write-up of this project, including the writing of the original draft.
Competing interests
The authors declare that there are no potential conflicts of interest concerning the research, authorship, and/or publication of this article.
Ethical approval
Before administration of the questionnaire, according to the Helsinki Declaration and APA ethical standards, employees (a) were advised about their right to decline or withdraw to participate at any time, (b) confirmed that the instructions were clear, (c) were informed about all relevant aspects of the study, and (d) agreed to participate in the study. Data were managed in line with the EU General Data Protection Regulation (GDPR); the study was approved by the ethics committee of the Faculty of Human and Social Sciences at the “Kore” University of Enna with code: UKE-IRBPSY-09.22.02.
Open access
