Technostress among nurses: boon or bane? The moderated mediation model

Abstract

Background

Today, increasing digitalization and intensified human-computer interaction in healthcare services have significantly changed nurses’ daily work processes. This situation has caused the emergence of a type of stress called technostress among nurses. The negative outcomes of technostress, generally referred to as its dark side, have been frequently emphasized in the literature. On the other hand, studies on the bright side of technostress, that is, its positive outcomes, are quite limited. Based on the job demands-resources model, this study assumes that technostress creators may have negative and positive outcomes.

Aim

In this context, this study aims to examine the indirect effects of technostress creators on job satisfaction through techno-eustress, on burnout through techno-distress, and how these indirect effects change depending on technical support and technology competence.

Methods

This study adopted a quantitative, correlational design and was conducted using a cross-sectional survey. Data were collected at a single time point from a convenience sample of nurses working in one hospital. The proposed research model was empirically tested using path analysis.

Results

Findings show that technostress creators increase burnout through techno-distress and job satisfaction through techno-eustress. It also shows that these indirect effects vary depending on the technology competence levels of nurses, and as the technology competence level increases, the indirect effects weaken.

Conclusion

In conclusion, the research findings revealed that, unlike other studies in the literature, technostress creators cannot be described as just a boon or a bane but can produce positive or negative outcomes through various mechanisms.

Introduction

Today, employees must cope with an almost unlimited number of stressors, both physical, such as noise, lack of sleep, and low blood sugar levels, on the one hand, and psychological, such as public speaking, social rejection, or as a recent stressor, human-computer interaction [1]. Technostress is a type of stress resulting from individuals’ psychological difficulties when using information and communication technologies [2]. The origins of technostress are based on psychological and biological disciplines and quantitative research in these areas. Current studies on technostress focus more on the effects of information technology use on individuals [3]. It is also stated that technostress is an experience that depends on employees’ characteristics, coping mechanisms, or adaptive capacities [4]. In this context, two sub-conceptual approaches stand out in the technostress literature. The first is techno-distress, which refers to the negative psychological stress that occurs when individuals perceive inadequacy, pressure, confusion, or threat in the face of technology. The other is techno-eustress, which is defined as the positive, motivating, and developmental stress type resulting from individuals’ interaction with technology [5]. Some psychologists also argue that certain stress levels help stimulate creative thinking, motivation, self-awareness, and productivity. However, excessive levels of emotional stress can be detrimental to human health in many ways [6]. Therefore, technostress must be addressed in both positive and negative aspects.

In recent years, there has been an increase in the number of studies on technostress, and its effects have mainly been examined on teachers, academics, and librarians [7]. However, its effects in the healthcare sector have been examined in limited numbers compared with other sectors [8, 9]. Countries such as China, India, the USA, the Netherlands, Estonia, and Türkiye are pioneers in digitalizing their hospitals [10]. This situation requires a closer examination of the relationship between healthcare professionals and technology. As healthcare professionals are increasingly confronted with digital technologies for clinical practice, interaction with patients, and administrative tasks, digitalization creates new tasks for healthcare professionals. It imposes tasks on them that are not directly part of their training [11]. As one of the main actors in the healthcare sector, nurses are also employees of a highly stressful profession, working under challenging conditions and striving to provide the best care to patients with limited staff and inadequate equipment. At the same time, they try to resolve conflicts within the team, meet the needs of patients and their relatives, and manage adverse emotional reactions [12].

Nursing is widely acknowledged as one of the most stressful professions, and numerous studies have examined various forms of stress experienced by nurses, including technostress. Meta-analyses have consistently shown that nurses report high levels of stress [13] and burnout [14], and there is also a growing body of research investigating their levels of technostress [8, 9]. While existing research has explored the effects of technostress creators (TSCs) on nurses’ occupational outcomes [15], relatively few studies have focused on the indirect pathways involving techno-distress and techno-eustress.

In addition, the role of variables that have the potential to influence this indirect effect, such as technical support and technology competence, has not been examined. The lack of studies in this direction has created a research gap, which serves as the main motivation for this study. Accordingly, the study pursues two main objectives. The first is to identify the indirect effects of TSCs on nurses’ job satisfaction and burnout through their perceptions of technostress. The second aim is to determine how these indirect effects may vary depending on nurses’ technology competencies and the technical support they receive. Accordingly, the study is expected to advance technostress research by investigating techno-eustress and techno-distress within the JD-R framework and by testing a moderated mediation model in the relatively underexplored healthcare context.

Theoretical background and hypotheses

The job demands-resources model

While the transactional model of stress and coping [16] recognizes that stressors can be appraised either as threats or challenges - and therefore may lead to either distress or eustress [17] - few studies have explicitly applied this model to understand techno-eustress in different contexts [5, 18, 19]. The transactional model’s focus on individual appraisal offers considerable insights into how personal evaluations determine whether a stressor functions as a challenge or a threat, which can lead to eustress or distress [17]. Despite the transactional model’s valuable emphasis on individual appraisal, the present study adopts the JD-R framework because it provides a more structured approach to classifying TSCs as either job demands or job resources, which aligns more directly with the dual conceptualization of techno-eustress and techno-distress. JD-R model is preferred due to its stronger explanatory power for categorizing stressors at the organizational level and for linking these categories more directly to specific work outcomes. Moreover, the JD-R model enables the incorporation of moderating variables such as technical support and technology competence, both of which are central to the present study.

The JD-R model divides job factors employees face into two main categories: job demands and job resources. Job demands are aspects of work that require sustained physical and psychological effort and are therefore associated with physiological or psychological costs. Such demands include excessive workloads, strict deadlines, conflicts with colleagues, and fear of losing one’s job. On the other hand, work resources encourage personal growth and alleviate the physiological and psychological costs associated with work. Examples such as social support, retaining control of work, and receiving performance feedback represent job resources [20]. According to the JD-R model, job demands do not automatically lead to negative outcomes. They become stressful when they are excessive or when employees lack sufficient job resources to manage their effects [10]. It is still debated whether the use of digital technology should be classified as a job demand or a resource [21]. In the logic of the JD-R model, TSCs can function either as job demands or job resources depending on how individuals perceive and respond to them. For instance, techno-overload—the perception of having to work faster and handle more information due to technology—typically acts as a job demand when it overwhelms nurses and depletes their energy. However, for some individuals, it can be perceived as a challenge that enhances productivity and provides a sense of accomplishment, thereby acting as a job resource. Similarly, techno-insecurity, which reflects concerns about losing one’s job due to technological changes, often functions as a psychological demand that increases stress levels. Yet, when accompanied by sufficient organizational support or seen as a motivation to upskill, it may serve as a resource that encourages professional development. These dual roles reflect the flexible nature of TSCs and highlight the importance of individual appraisal and contextual factors in determining their impact. In this context, a study on doctors revealed that digital health technologies serve as a job resource rather than a job demand for doctors [22]. However, as limited research has examined nurses in this context, this study was designed to incorporate both techno-distress and techno-eustress variables into the research model. Techno-eustress is defined as a positive form of stress that occurs when individuals perceive information systems as challenges that provide opportunities for learning, skill development, and improved job performance, whereas techno-distress refers to the negative stress experienced when these systems are appraised as overwhelming threats. Accordingly, this study is expected to provide a deeper understanding of how TSCs influence both positive and negative work outcomes.

Technostress creators, burnout, and job satisfaction

Technostress is the stress people experience due to using information and communication systems and technologies [5]. TSCs were proposed by Tarafdar et al. [23] and consisted of techno-overload, techno-invasion, techno-complexity, techno-uncertainty, and techno-insecurity. Although analyzed collectively, these dimensions generally function as job demands in the JD-R framework because they require sustained cognitive and emotional effort and may lead to energy depletion and strain [24]. However, under specific conditions—such as when organizational or personal resources are high—certain creators (e.g., techno-overload or techno-complexity) may be appraised as challenges rather than hindrances, thereby acting as job resources that stimulate learning, skill development, and personal growth [25].

Burnout, another variable in this study, is a long-term response to chronic emotional and interpersonal stressors in the workplace. Burnout is considered one of the most prevalent outcomes of the health-impairment process in the JD-R model. In this context, this study focuses on the emotional exhaustion and cynicism dimensions of burnout, which are regarded as its fundamental components [26]. Emotional exhaustion reflects the depletion of physical and psychological resources due to excessive job demands, whereas cynicism denotes a psychological withdrawal and detachment from work. Although reduced professional efficacy is also part of the broader burnout construct, JD-R studies generally emphasize energy depletion and disengagement as the primary consequences of prolonged exposure to high job demands [27]. Burnout remains a social problem in the service sector that requires close relationships with people [28]. High levels of burnout among nurses negatively affect their work performance, job satisfaction, and overall quality of working life [29–31]. Job satisfaction is the degree to which individuals feel positive or negative about their job. In other words, it is the attitude or emotional response to one’s duties and the physical and social conditions of the workplace. Job satisfaction among healthcare professionals is increasingly recognized as an essential determinant as it directly affects any country’s health system [32]. In the literature, the direct impact of technostress on positive work outcomes, such as job satisfaction, and negative work outcomes, such as burnout, has been examined in the healthcare sector and various other sectors. Research results reveal that technostress decreases positive work outcomes, such as job satisfaction, and increases negative work outcomes, such as burnout [15, 33, 34]. However, although some studies, such as Califf et al. [35], have explored the indirect effect of technostress on job satisfaction through both techno-distress and techno-eustress perceptions, the indirect effects of technostress on positive and negative work outcomes through various variables remain relatively underexplored.

Techno-distress as a mediator

State-of-the-art medical devices that integrate with information technologies can support nurses’ clinical decision-making processes and help them provide better-quality healthcare services. However, these devices can also increase personal stress levels, limit nurses’ clinical autonomy, and increase administrative workload due to the complexity of digital interfaces, the need for constant system monitoring, and data entry [36]. In the logic of the JD-R model, such technology-related factors act as job demands when they require sustained cognitive and emotional effort and lead to physiological or psychological costs [24, 37]. Techno-distress, which emphasizes the challenging aspect of technology, is a phenomenon that embodies the negative stress that individuals face when using information technologies [5]. From a JD-R perspective, techno-distress represents the health-impairment process, in which excessive job demands drain energy and psychological resources, increasing the risk of burnout [27]. At this point, how a person chooses to react psychologically to a stressful situation caused by technology comes into play, and this reaction is shaped by how they perceive the stressful situation [15]. From the perspective of nurses, many reasons, such as the necessity of using health information technologies in order to ensure continuity of care and therefore spending more time, the threat of losing the job to someone who knows the technology better, the constant changes in technologies and the uncertainty caused by this, and the perception of health technologies as very complex for health service providers, cause employees to develop a negative perception of technology [38, 39]. Due to the characteristics of the nursing profession, nurses may experience different degrees of burnout [31]. Various complex factors lead to burnout in nurses. Recently, negative attitudes toward using information systems have been emphasized as factors that increase employee burnout [5]. Consistent with the JD-R health-impairment pathway, when such TSCs function as job demands and are perceived predominantly as hindrances rather than challenges, they are expected to increase techno-distress, which in turn depletes emotional and physical resources and contributes to burnout. In line with these arguments, we hypothesize as follows:

H₁

Techno-distress mediates the effect of TSCs on burnout.

Techno-eustress as a mediator

Selye [40] first discussed the concept of eustress. This concept is related to the Yerkes-Dodson Law, which states that increased stress can benefit people’s performance until an optimum level is reached. Those who experience a certain level of stress may be more productive and efficient than those who do not [41]. In this context, technostress is a good stress with positive outcomes [42]. Nelson and Simmons [43] described the relationship between techno-distress and techno-eustress using the analogy of balancing hot and cold water flowing into a bathtub. Accordingly, techno-distress represents cold water, while techno-eustress represents hot water, and an ideal temperature can only be achieved by appropriately combining these two elements. Thus, techno-stress is neither wholly positive nor completely negative until evaluated by the individual; stress factors are initially considered neutral and only become positive or negative according to the individual’s perception. Individuals are subject to stress factors in the process of self-evaluation. This evaluation process may lead to either techno-eustress, which tends to be associated with positive psychological responses and beneficial outcomes, or techno-distress, which is generally linked to negative reactions and adverse outcomes. During the process, it is critical to understand which events can lead to positive or negative emotional states and how and why individual and organizational consequences can occur [44]. The finding that nurses using digital hospital systems in the healthcare sector are less mentally exhausted and less prone to medical errors is interesting in this context. Various studies have examined the positive effects of digital technologies on business life [45]. Quite a few studies examining the positive aspects of technostress have revealed that technostress increases job satisfaction [15, 18]. From the perspective of the JD-R model, such technology-related stressors can function as job resources when they stimulate personal growth, learning, and development, thereby activating the motivational process that leads to positive outcomes such as work engagement or job satisfaction [24, 27]. When TSCs are perceived as challenges that enhance efficiency, competence, and task mastery, they are expected to increase techno-eustress, which in turn fosters job satisfaction by satisfying psychological needs and promoting intrinsic motivation. Therefore, in this study, to expand existing technostress research, techno-eustress was considered as a tool in the model, and the following hypothesis was developed:

H₂

Techno-eustress mediates the effect of TSCs on job satisfaction.

Technology competence as a moderator

The concept of technology competence is closely related to technological capabilities and skills. Technology competence can be considered a result of developing specific technical skills and innovations [46]. Technology competencies in the healthcare sector can be considered, together with the ever-changing nature of the equipment used. The requirements of technical equipment, such as ease of use, smaller size, and longer lifespan, have led to more complex software programming and device designs. According to this perspective, medical devices are systems consisting of mechanical, electrical, and chemical subsystems rather than independent products [47]. One of the health professionals who actively uses these systems is nurses. Rapid technological change in hospitals and health systems since the 1950s has forced nurses to learn digital technology applications and constantly update their digital skills to maintain their clinical competence [48]. While this transformation creates technical problems and high performance demands that increase workload in some cases [49], it can also have positive effects that support the development of digital skills, reinforce professional competence, or increase individual motivation. The limited technology competencies of nurses and the poor functioning of electronic health records in the healthcare sector increase time pressure and pave the way for the formation of distress [50]. However, users with high technology self-efficacy may have the ability to manage additional work caused by disruptions in their access to information [51]. From the perspective of the JD-R model, technology competence can be conceptualized as a personal resource that facilitates goal achievement, stimulates learning and development, and helps individuals cope with job demands [24, 27]. As JD-R theory predicts, personal resources can buffer the negative effects of high job demands on strain (health-impairment process) and simultaneously amplify the positive effects of job resources on motivational outcomes (motivational process). Accordingly, high technology competence is expected to weaken the relationship between TSCs conceptualized as job demands and techno-distress, as well as reduce the indirect impact of TSCs on burnout through techno-distress. Conversely, when TSCs are perceived as job resources (e.g., challenges stimulating skill development), high technology competence is expected to strengthen the link between TSCs and techno-eustress and, consequently, its indirect effect on job satisfaction. The research hypotheses formed in this direction are as follows.

H₃

Technology competence moderates the effect of TSCs on techno-distress, such that this effect is weaker when technology competence is high.

H₄

Technology competence moderates the indirect effect of TSCs on burnout such that this indirect effect is weaker when technology competence is high.

H₅

Technology competence moderates the effect of TSCs on techno-eustress such that this effect is stronger when technology competence is high.

H₆

Technology competence moderates the indirect effect of TSCs on job satisfaction, such that this indirect effect is stronger when technology competence is high.

Technical support as a moderator

Technical support is an organizational mechanism that has the potential to reduce the negative impact and level of technostress [2]. As new digital technologies are often introduced quickly, end users may need training and guidance on how to use new systems, especially in the adaptation process. Support can help change end users’ perspectives on technology. Therefore, technical support is a situational factor that can reduce the impact of healthcare workers’ perceptions and response choices in stressful situations [15]. In other words, technical support refers to a technical support system or team that responds positively and quickly to nurses’ requests or doubts about any technological device. The negative effects of technostress can be reduced through these support mechanisms. Nurses also use complex devices and systems such as hospital information systems, infusion pumps, ultrasound devices, and heart monitors to deliver healthcare services and serve patients by following strict care policies and procedures [52]. In this context, both organizations and employees welcome technical support because it benefits both parties. For example, technical support enables employees to perform their tasks easily and faster [51]. From the JD-R perspective, technical support is an organizational resource that facilitates goal attainment, reduces job demands, and stimulates personal growth, thus influencing both the health-impairment and motivational processes [24, 27]. JD-R theory posits that organizational resources buffer the negative impact of high job demands on strain while amplifying the positive effects of job resources on motivational outcomes. Accordingly, high technical support is expected to weaken the relationship between TSCs conceptualized as job demands and techno-distress and to reduce the indirect impact of TSCs on burnout through techno-distress. Conversely, when TSCs are perceived as job resources (e.g., challenges stimulating skill development), high technical support is expected to strengthen the link between TSCs and techno-eustress and, consequently, its indirect effect on job satisfaction. In this context, we argue that technical support offered in healthcare organizations can not only reduce negative perceptions of technology but also increase positive perceptions. In this context, we hypothesize the following:

H₇

Technical support moderates the effect of TSCs on techno-distress, such that this effect is weaker when technical support is high.

H₈

Technical support moderates the effect of TSCs on techno-eustress, such that this effect is stronger when technical support is high.

H₉

Technical support moderates the indirect effect of TSCs on burnout, such that this indirect effect is weaker when technical support is high.

H₁₀

Technical support moderates the indirect effect of TSCs on job satisfaction, such that this indirect effect is stronger when technical support is high.

Building on these hypotheses, the proposed research model is depicted in (Fig. 1.)

Fig. 1.

Research model with hypothesized relationships

Methods

Study design and setting

This study was conducted to examine the positive (techno-eustress) and negative (techno-distress) effects of TSCs resulting from nurses’ interaction with technology and to reveal the indirect effects of these effects on burnout and job satisfaction. In addition, the moderating effects of technology competence and technical support on these relationships were investigated. For this purpose, the research was conducted within the framework of the correlational research design, one of the quantitative research methods, and with a cross-sectional design. To test the directional relationships between the variables, structural equation modelling (SEM) was used, and mediation and moderated mediation analyses were conducted. The research design was preferred because it allows for testing complex relationships between the variables and is suitable for working with data based on subjective perceptions.

The study was conducted in a 400-bed public training and research hospital in the Central Anatolia Region of Turkey, where technology is used intensively in clinical practices. This setting was selected as an appropriate context for the study due to its strong technological infrastructure and the constant interaction of nurses with various health technologies in their daily workflow.

Participants

A total of 330 nurses working in a public hospital took part in this study. Most participants were women (81.4%, n = 263); ages ranged from 21 to 55 (mean = 36.8, sd = 8.5), 36% worked only during the day shift, and the remainder worked both day and night shifts. Participants had been working at the hospital for a minimum of 1 month and a maximum of 36 years (mean = 14.5 years, sd = 7.2 years) and cared for an average of 20.5 patients daily (sd = 30.7). Since seven participants had at least 5% missing data in their responses, they were excluded from the analysis. The final sample consisted of 323 participants with complete data.

Measures

Technostress creators: TSCs were measured using the Technostress Creators Scale developed by Ragu-Nathan et al. [2], which consists of 23 items. A sample item from this scale is: “I need a long time to understand and use new technologies.” Cronbach’s alpha coefficient (CA) for the overall scale was 0.83. However, this study did not include items in the techno-invasion subscale. The main reason for this is that nurses, unlike office workers, work in shifts and do not tend to use health information technologies at home or for personal purposes [15]. Participants responded to the scale items on a five-point Likert scale ranging from “strongly disagree” to “strongly agree.”

Technical support: Technical support, one of the subscales of the Technostress Inhibitors Scale developed by Ragu-Nathan et al. [2], consists of four items. For instance, one item reads: “Our IT help desk is easily accessible.” The subscale’s CA was 0.92. Participants responded to the scale items on a five-point Likert scale ranging from “strongly disagree” to “strongly agree.”

Technology competence: The technology competence scale was developed by Tarafdar et al. [53]. There are four items in this scale. An illustrative item is: “I can use this technology to improve my productivity.” The CA of the scale was 0.94. Participants responded to the scale items on a five-point Likert scale ranging from “strongly disagree” to “strongly agree.”

Techno-eustress: The techno-eustress scale was created by Califf et al. [15] by adapting the scale developed by O’Sullivan [41] to measure eustress in the context of hospitals and technology. Although Califf et al. [15] did not report internal consistency statistics, we assessed the reliability of the 10-item scale in our sample. The CA was 0.87. An item reflecting this construct is: “How often do you feel that stress at work that stems from technology has a positive effect on your performance?” Participants responded to the scale items with a six-point Likert scale ranging from “never” to “always.” High scores on this scale indicate that stress caused by technology is perceived more positively. In contrast, low scores indicate that stress caused by technology is perceived less positively and do not indicate the presence of distress at all.

Techno-distress: The Techno-distress Scale was created by Califf et al. [15] by adapting the scale developed by Kessler et al. [54] to measure distress in the context of hospitals and technology. One of the items is phrased as follows: “During the last 30 days, about how often did you feel hopeless about your hospital’s technology?” The CA of the ten items was 0.95. Participants responded to the items on a 5-point Likert-type scale ranging from “never” to “always.” High scores on this scale indicate an increased pattern of distress caused by health information technology. Conversely, low scores on this scale indicate a low pattern of distress caused by health information technology and do not indicate the presence of eustress at all.

Job satisfaction: The scale was developed as 18 items by Brayfield and Rothe [55] and converted into a short form with five items by Judge et al. [56]. This short form was adapted into Turkish by Başol and Çömlekçi [57]. An example statement included in the short form is: “I do my job with pleasure.” The CA of the scale was 0.90. Participants responded to the items on a 5-point Likert-type scale ranging from “strongly disagree” to “strongly agree.”

Burnout: The burnout scale was developed as 21 items by Pines and Aronson [58] and converted into a short form with ten items by Malach-Pines [59]. This short form was adapted into Turkish by Capri [60]. One representative item is: “Disappointed with people.” The CA of the scale was 0.94. Participants responded to the items with a 7-point Likert scale ranging from “never” to “always.”

A demographics questionnaire included gender, age, experience, number of patients, and nursing shift. TSCs, techno-distress, techno-eustress, technical support, and technology competence scales were originally developed in English. Therefore, the scales were translated into Turkish by the researchers and then back-translated into English by two independent bilingual experts. This process was carried out in accordance with the translation–back-translation method suggested by Brislin [61] and aimed to ensure linguistic equivalence. After checking the accuracy and understandability of the translations obtained, they were evaluated in terms of content appropriateness and cultural validity by a panel of three academicians specialised in health informatics and organizational behavior. After making the necessary minor corrections according to expert opinions, a pilot application was carried out. In this context, the researchers carried out the translation process of the scales used in the study.

In addition to the translation and expert review process, the psychometric properties of the Turkish versions of the scales were examined during the main data analysis. Construct validity was assessed through exploratory factor analysis (EFA), which confirmed that items loaded appropriately onto their intended factors. Furthermore, internal consistency was supported by CA above acceptable thresholds (see Table 1).

Table 1.

Descriptive statistics and correlations of the study variables

Variables	M	SD	1	2	3	4	5	6	7	8	9	10
1.TSCs	2.60	0.55	0.83
2.TD	2.47	0.93	0.43**	0.94
3.TE	2.24	0.84	–0.02	–0.25**	0.87
4.BO	3.75	1.42	0.29**	0.45**	0.03	0.94
5.JS	3.38	0.96	–0.12*	–0.15**	0.11*	–0.44**	0.90
6.TSP	3.27	0.89	–0.17**	–0.29**	0.31**	–0.11	0.23**	0.91
7.TC	3.56	0.90	–0.14*	–0.10	0.10	–0.28**	0.34**	0.26**	0.93
8.NOP	20.53	30.74	–0.09	–0.05	–0.03	–0.13*	–0.01	–0.02	0.06	–
9.EXP	175.5	109.7	0.01	–0.09	–0.01	–0.20**	0.11*	0.05	0.05	0.14*	–
10.AGE	36.82	8.46	0.02	–0.06	–0.01	–0.18**	0.12*	0.03	0.02	0.011*	0.93*	–

Notes: M: Mean; SD: Standard Deviation; TSCs: Technostress creators; TD: Techno-distress; TE: Techno-eustress; BO: Burnout; JS: Job satisfaction; TSP: Technical support; TC: Technology competence; NOP: Number of patients (in day); EXP: Experience (in months); Age (in years); Diagonals (in bold) represent Cronbach’s alpha coefficient; *p < 0.05 **p < 0.01 (two-tailed)

Procedure

The institutional review board of the authors’ university approved the study on 29.10.2022, numbered E-86837521-050.99-299436. The convenience sampling method was used in the hospital setting. In selecting participants, the criterion that participants should use at least one of the hospital information systems was considered, and this was notified to the participants before participation. Then, the researchers conducted the survey face-to-face. Before the data collection, the participants were given an informed consent form and asked to read and approve it. Participants who read and approved the informed consent form were included in the study.

Statistical analysis

We used SPSS (Version 23.0) for descriptive statistics and correlations. We evaluated factorial validity by conducting an EFA using SPSS software to examine the underlying structure of the constructs and to provide preliminary evidence of construct validity. Reliability was also assessed with SPSS by calculating CA. We also used AMOS (Version 23.0) to test the fit of the research model and examine the research hypotheses. In addition to testing the moderated mediation hypotheses/the conditional indirect effect, we calculated the index of moderated mediation (IMM) suggested by Hayes [62] in the AMOS program. As in mediation analyses, the sampling distribution will be irregular in form because the IMM is a product of two regression coefficients. For this reason, we used bootstrap confidence intervals when making inferences about IMM. In this context, we preferred 5,000 bootstraps and 95% bias-corrected confidence intervals (CI). The absence of a zero value between the lower and upper limits of the bootstrap confidence intervals indicates that the relevant effect is statistically significant.

Results

Common method bias test

Since all data were collected from the same participants through self-reported questionnaires, there was a potential risk of common method bias, which could affect the validity of the findings. Before moving on to the data analysis phase, Harman’s single-factor test, recommended by Podsakoff et al. [63], was used to determine whether common method bias would affect the data. Researchers load all items into EFA and examine the unrotated factor solution in this method. The results showed that the first main factor explained only 21.57% of the total variance. This value is well below the threshold value (50%) and shows that common method bias does not have a worrying effect on our results [64].

Nevertheless, as noted by Podsakoff et al. [63], Harman’s test alone may not be sufficient to rule out method bias. Accordingly, we implemented several procedural remedies to minimize such bias. These included: careful wording of the items during the translation and adaptation process; assurances of anonymity and confidentiality; explicit instructions that there were no right or wrong answers; and the physical distribution and collection of the questionnaires in sealed envelopes using a secure drop-box. These precautions support the conclusion that common method bias does not pose a serious threat to the validity of our findings.

Factorial validity and reliability analysis results

We examined CA to evaluate the reliability of the structures. As seen in Table 1, CAs for the variables are above 0.70. These results show that all structures have sufficient reliability. Additionally, we performed EFA using the principal axis factoring (PAF) extraction method and Promax rotation, an oblique rotation technique to test the factorial validity. PAF was chosen instead of principal components analysis because it is more appropriate for identifying latent constructs by focusing on common variance rather than total variance. Promax, an oblique rotation method, was used since it allows for correlation among factors, which is consistent with the expectation that psychological constructs such as technostress, burnout, and job satisfaction are interrelated [65].

Before performing EFA, Bartlett’s sphericity test was performed, and the Kaiser-meyer-olkin (KMO) value was examined to evaluate whether the data were suitable for factor extraction. In the factor analysis with ten factors, the KMO value was found to be 0.88, and the Bartlet sphericity test result was significant (p < 0.001). The findings revealed that the data were suitable for factor analysis [66]. As a result of the EFA, all items are loaded under their factors. Only the sixth and seventh items of the techno-eustress scale were removed due to low factor loadings (0.08 and 0.10, respectively). Interestingly, all the excluded items were the only reverse-coded ones, suggesting that negative phrasing may have impaired their performance due to possible respondent confusion. The factor loading values of the remaining items of the scales are above 0.40. A total of ten factors explains 62.2% of the total variance. Therefore, the findings indicate that the scales used in the current study are reliable and demonstrate acceptable factorial validity, supporting their suitability for the research context.

Descriptive statistics and correlations

The mean, standard deviation, and zero-order correlations for the variables (including sociodemographic variables) are presented in Table 1. Main findings of the study showed that burnout was positively associated with TSCs and techno-distress (r = 0.29, p < 0.01; r = 0.45, p < 0.01) and negatively related to technology competence (r = 0.28, p < 0.01). Moreover, job satisfaction was negatively correlated with TSCs and techno-distress (r = –0.12, p < 0.05; r = –0.15, p < 0.01) and positively related to techno-eustress (r = 0.11, p < 0.05).

Testing for the proposed model

Structural equation modelling was used to test the path model (Fig. 2) created with observed variables.

Fig. 2.

The path analysis model. Note: The double-headed arrows between techno-distress and techno-eustress and between burnout and job satisfaction represent the covariances between the disturbances of the relevant variables. Unstandardized path coefficients are reported. *p < 0.05, **p < 0.01, *** p < 0.001

As a result of the model test, the sufficient level of model fit indices (χ²(8) = 15.09, p > 0.05, χ²/df = 1.89, CFI = 0.99, RMSEA = 0.05, SRMR = 0.02) that the collected data fits the proposed model well. In addition, research hypotheses were tested, and the results are given in Table 2.

Table 2.

Mediation and moderated mediation analysis results

	b	SE	LLCI	ULCI	Hypotheses results
Outcome: TD (R2 = 0.271)
TSCs	0.942**	0.361	0.067	1.706
TC	0.584**	0.200	0.151	1.067
TSP	−0.581**	0.220	−1.239	−0.030
TSCs x TC	−0.213**	0.075	−0.394	−0.046	H3 supported
TSCs x TSP	0.135	0.084	−0.073	0.382	H7 rejected
Age	0.009	0.015	−0.025	0.044
NOP	0.001	0.002	−0.004	0.004
Experience	−0.001	0.001	−0.004	0.001
NS	0.191*	0.100	0.006	0.389
Outcome: BO (R2 = 0.285)
TD	0.638***	0.069	0.488	0.788
Age	−0.013	0.022	−0.058	0.034
NOP	−0.003	0.002	−0.006	0.001
Experience	−0.001	0.002	−0.004	0.003
NS	0.045	0.150	−0.245	0.318
Bootstrapping results of the indirect effects
TSCs → TD → BO	0.601	0.280	0.061	1.171	H1 supported
Outcome: TE (R2 = 0.136)
TSCs	0.971**	0.357	0.073	1.704
TC	0.480*	0.199	0.040	0.985
TSP	0.520*	0.218	−0.047	1.119
TSCs x TC	−0.178*	0.074	−0.368	−0.017	H5 partially supported
TSCs x TSP	−0.095	0.083	−0.313	0.127	H8 rejected
Age	−0.005	0.015	−0.035	0.024
NOP	−0.001	0.001	−0.004	0.001
Experience	0.001	0.001	−0.002	0.002
NS	−0.235*	0.099	−0.448	−0.039
Outcome: JS (R2 = 0.159)
TE	0.116*	0.057	0.004	0.242
Age	0.020	0.017	−0.020	0.052
NOP	−0.001	0.002	−0.001	0.002
Experience	−0.001	0.001	−0.001	0.002
NS	−0.015	0.113	−0.015	0.210
Bootstrapping results of the indirect effects
TSCs → TE → JS	0.113	0.075	0.006	0.321	H2 supported
Bootstrapping results for the conditional indirect effects
IMM1 (TSCs → TD → BO at different levels of TC)	−0.136	0.058	−0.265	−0.031	H4 supported
IMM2 (TSCs → TD → BO at different levels of TSP)	0.086	0.073	−0.043	0.252	H9 rejected
IMM3 (TSCs → TE → JS at different levels of TC)	−0.021	0.016	−0.068	−0.001	H6 partially supported
IMM4 (TSCs → TE → JS at different levels of TSP)	−0.011	0.015	−0.057	0.009	H10 rejected

Notes: M: Mean; SE: Standard error; TSCs: Technostress creators; TD: Techno-distress; TE: Techno-eustress; BO: Burnout; JS: Job satisfaction; TSP: Technical support; TC: Technology competence; NOP: Number of patients (in day); EXP: Experience (in months); Age (in years); NS: Nursing shifts (0 = Only day shift, 1 = Day and night shifts); LLCI: Lower limit confidence interval; ULCI: Upper limit confidence interval; Unstandardized effects (b) are reported in the table; Bootstrap resamples = 5,000; *p < 0.05 **p < 0.01, ***p < 0.001 (two-tailed)

Results show that TSCs have a statistically significant and positive effect on techno-distress (b = 0.94, p < 0.01) and techno-eustress (b = 0.97, p < 0.01). Additionally, techno-distress has a positive effect on burnout (b = 0.64, p < 0.001), and techno-eustress has a statistically significant and positive effect on job satisfaction (b = 0.12, p < 0.05).

The bias-corrected bootstrap confidence interval findings in Table 2 show that TSCs have a statistically significant positive indirect effect on burnout through techno-distress (b = 0.60, 95% CI= [0.06, 1.17]). Additionally, TSCs positively affect job satisfaction through techno-eustress (b = 0.11, 95% CI= [0.01, 0.32]). Therefore, the results support H₁ (Techno-distress mediates the effect of TSCs on burnout) and H₂ (Techno-eustress mediates the effect of TSCs on job satisfaction).

We examined the moderating role of technology competence and technical support. Technology competence has a moderating role in the relationship between TSCs and techno-distress (b = − 0.21, p < 0.01) and the relationship between TSCs and techno-eustress (b = − 0.18, p < 0.05). The fact that the moderating effect is negative in both cases shows that as people’s technology competence level increases, the effect of TSCs on both techno-distress and techno-eustress weakens. These results support H₃ (Technology competence moderates the effect of TSCs on techno-distress, such that this effect is weaker when technology competence is high). They also provide partial support for H₅ (Technology competence moderates the effect of TSCs on techno-eustress such that this effect is stronger when technology competence is high), as the moderation effect, although statistically significant, was contrary to our expectation. Specifically, the positive effect of TSCs on techno-eustress became weaker—not stronger—at higher levels of technology competence. Technical support does not have a moderating role in the relationship between TSCs and techno-distress, nor in the relationship between TSCs and techno-eustress, because CI contains the value zero. For this reason, H₇ (Technical support moderates the effect of TSCs on techno-distress, such that this effect is weaker when technical support is high) and H₈ (Technical support moderates the effect of TSCs on techno-eustress, such that this effect is stronger when technical support is high) were rejected.

As mentioned above, IMM was calculated to examine the conditional indirect effect of TSCs on both burnout and job satisfaction. In this context, technology competence moderated the indirect effect of TSCs on burnout via techno-distress (b = − 0.14, 95% CI= [− 0.27, − 0.03]). IMM₁ has a negative value, which means that the indirect effect is negatively related to the moderator. Therefore, as the technology competence level of nurses increases, the indirect effect of TSCs on burnout via techno-distress weakens. Also, technology competence moderated the indirect effect of TSCs on job satisfaction via techno-eustress (b = − 0.02, 95% CI= [− 0.07, − 0.01]). Similar to the previous finding, as nurses’ technology competence level increases, the indirect effect of TSCs on job satisfaction via techno-eustress weakens. These findings support H₄ (Technology competence moderates the indirect effect of TSCs on burnout such that this indirect effect is weaker when technology competence is high). They also offer partial support for H₆ (Technology competence moderates the indirect effect of TSCs on job satisfaction, such that this indirect effect is stronger when technology competence is high), since the conditional indirect effect weakened, rather than strengthened, as technology competence increased—opposite to what we had hypothesized. When the moderating role of technical support on indirect effects is examined, it can be seen in Table 2 that technical support does not have a statistically significant moderating effect on both indirect effects. For this reason, H₉ (Technical support moderates the indirect effect of TSCs on burnout, such that this indirect effect is weaker when technical support is high) and H₁₀ (Technical support moderates the indirect effect of TSCs on job satisfaction, such that this indirect effect is stronger when technical support is high) were rejected.

Discussion

Recent years have witnessed a growth in scholarly interest in technostress [5, 67], often associated with adverse outcomes. In this sense, a plethora of studies previously attempted to unearth the “dark side” of technostress [7, 38, 67, 68]. Despite few, the literature also hosts research investigating the “bright side” of technostress, predicating the idea that technostress may bear desirable aspects [15, 18, 19, 35, 69]. Ultimately, the present study explored the outcomes when technostress, often pronounced by its adverse aspects and outcomes, is perceived positively (techno-eustress) or negatively (techno-distress). It also investigated the roles of technical support and technology competence in this process. However, due to the cross-sectional nature of the study and the sample data obtained from a single institution, the findings should be interpreted with caution. These methodological limitations restrict both the testing of causal relationships and the generalisation of the results to broader populations.

One remarkable finding of this study is that TSCs increase nurses’ burnout through techno-distress. Similarly, a substantial body of research in the literature sought TSCs’ direct impacts on burnout and similar negative outcomes [70, 71]. Studies examining the direct effect of TSCs on burnout in various areas reveal that there is a positive effect [72, 73]. In line with these findings, our study also suggests that TSCs may be indirectly related to burnout through techno-distress. In addition, Califf’s et al. [35] findings demonstrated that TSCs affected their intention to leave, a negative outcome for nurses similar to burnout, through techno-distress. When all these findings are considered together, they suggest that techno-distress may be an important mechanism in explaining the potential effect of TSCs on burnout.

Another important finding of our study indicates that TSCs can increase nurses’ job satisfaction through techno-eustress. The relationships between TSCs and positive job outcomes (e.g., job satisfaction) were consistently explored, particularly among nurses in healthcare and other employment fields. The findings of these studies demonstrated that TSCs are negatively associated with job satisfaction [73–75]. However, the potential role of certain mediating mechanisms (e.g., techno-eustress) in explaining the relationship between these variables has not been sufficiently explored in previous studies. Contrary to the mentioned findings, we could reveal that TSCs increased job satisfaction through techno-eustress among participating nurses. This finding is consistent with those reported by Nascimento et al. [18, 19], who highlighted the positive influence of techno-eustress on job satisfaction. Therefore, techno-eustress - one is a positive perception of technology-causing stress - can be interpreted as a meaningful variable in understanding the relationship between TSCs and job satisfaction.

Furthermore, we concluded that a high level of technology competence reduced TSCs’ direct positive effect on techno-distress and indirect positive effect on burnout. Regarding the ability to understand and utilize technology and the level of technological knowledge, technology competence secures a more convenient and efficient use of technology, helping reduce the distress caused by TSCs. In their study on the relationship between digital competence and technostress in healthcare workers, Golz et al. [11] found that higher digital competence is significantly associated with lower technostress; this finding overlaps our results. In addition, we concluded that a higher level of technology competence decreased TSCs’ direct positive effect on techno-eustress and indirect positive effect on job satisfaction. We also conclude that higher levels of technology competence reduce the direct positive effect of TSCs on techno-eustress and the indirect positive effect on job satisfaction. Although technology competence moderates both the direct and indirect effect, the direction of this effect was opposite to our expectations. However, no studies directly supporting our findings have been found in the literature. Some studies suggest that the relationships between TSCs and outcomes may be non-linear, for example, in an inverted U-shape [76]. Such non-linear relationships may also lead to the fact that a certain level of technology competence may increase techno-eustress, while a higher level may have a different effect. This may be explained by the fact that individuals with high technology competence tend to perceive technology as an ordinary part of their work and therefore may be less likely to notice its positive effects. These findings suggest that the effects of technology competence are not always positive and may have a complex interaction structure. The partial support for hypotheses H₅ and H₆ in our study is similar to these exceptional results.

Finally, our findings showed that technical support did not have a moderating role in TSCs’ direct effects on techno-distress and techno-eustress and indirect effects on burnout and job satisfaction. Technical support - an organizational strategy for coping with technostress and its adverse consequences - is considered among situational factors to reduce the impacts of one’s perception of and response to a stressful situation [15]. Also, we concluded that technical support has a positive effect on techno-eustress and a negative effect on techno-distress. Zhao et al. [77] study showed the positive impacts of technical support on techno eustress. However, the absence of a moderating effect of technical support in our model suggests that technical support may not play the expected role in altering the relationship between TSCs and techno-distress or techno-eustress.

Theoretical implications

We believe our findings have a robust potential to contribute to understanding the association between TSCs and their positive and negative consequences, such as job satisfaction and burnout. Although a considerable number of studies previously addressed the relationships of TSCs with job satisfaction and burnout, the present research accomplishes bringing a distinct perspective to the subject by demonstrating that TSCs can exert different impacts on job satisfaction and burnout through various related mechanisms such as techno-eustress and techno-distress. In particular, our findings support the argument that TSCs can also lead to positive outcomes through techno-eustress, even though they are often associated with adverse consequences in the literature. Thus, instead of encapsulating TSCs in negative connotations, this study revealed that they are likely to produce desirable results when perceived positively.

Moreover, another distinct contribution of our research may lie under the finding that the indirect effects of TSCs on positive and negative job-related outcomes vary by technology competence. From a theoretical perspective, this study contributes to the JD-R framework by clarifying the conditions under which TSCs operate as job demands or as job resources. Consistent with the JD-R model’s dual pathways, our findings demonstrate that TSCs trigger the health-impairment process (via techno-distress) when they are appraised as overwhelming and uncontrollable, depleting energy and increasing burnout, whereas they activate the motivational process (via techno-eustress) when perceived as manageable challenges supported by sufficient resources, thereby enhancing job satisfaction. By explicitly showing how the same TSCs can simultaneously function as demands and resources depending on individual appraisal and contextual factors (e.g., technology competence, organizational support), this research extends the JD-R model to the domain of technostress in healthcare.

Additionally, our study uses the JD-R theory to offer a fresh perspective on how TSCs impact healthcare workers’ job satisfaction and burnout. This broadens our understanding of the complex relationship between TSCs and job-related outcomes in healthcare. Finally, this study offers a helpful framework to guide prospective research on technostress, techno-eustress, techno-distress, and job-related outcomes in healthcare institutions.

Practical implications

TSCs have positively affected job satisfaction through techno-eustress. This finding is particularly interesting for employees in a dynamic and complex work environment such as healthcare institutions. In this context, positive stress can be considered an alternative motivation tool to increase the job satisfaction of nurses. Therefore, technology can be integrated into employees’ work goals, improving job satisfaction and productivity. In addition, workshops, simulations, and on-the-job training can be organized for employees to simplify complex technologies in the organization and enable them to improve their work routines. Additionally, the findings show that TSCs influence employee burnout through techno-distress and that technology competence weakens this effect. Therefore, the institution can offer training programs to improve nurses’ technology competence.

Limitations and future research directions

In addition to its important contributions, this study has some limitations. First, due to the study’s cross-sectional design, each variable was measured simultaneously without temporal priority. Moreover, causality in the study design hinges on previous findings and theories; therefore, one may need to be rigorous when drawing causal inferences from our findings. This situation prevents the establishment of causal relationships between variables and limits the interpretation of mediation analyses in particular. Mediation analyses assume that the independent variable precedes the mediator variable in time, and the mediator variable precedes the dependent variable in time. However, the presence of only a single measurement time point in our study makes it impossible to test this assumption empirically. Therefore, the mediation effects reported in the study should be considered exploratory rather than confirmatory. In addition, future research with a longitudinal research design may help verify our findings. Although our sample size provided sufficient power for statistical analyses, the analysis of the data gathered from a single hospital and professional group cannot secure the generalizability of our findings to all healthcare professionals. Finally, we had to employ convenience sampling, one of the non-probability sampling methods in this study, due to financial, time-related, and environmental limitations, which may restrict our findings’ generalizability and external validity. Additionally, the inclusion of four technostress constructs (techno-overload, techno-complexity, techno-uncertainty, and techno-insecurity) into a single composite variable is a potential limitation. This limitation may overlook the differential effects of each sub-dimension on techno-distress and techno-eustress. When these limitations are considered together, it is clear that the research findings must be interpreted with caution. Therefore, future studies designed using longitudinal, multi-centre, and probability-based sampling methods, which also examine the sub-dimensions of technostress separately, are important in terms of both testing the validity of the current findings and obtaining more generalisable results.

Conclusion

The present study may be considered the new effort to understand TSCs’ indirect effects on burnout and job satisfaction through techno-distress and techno-eustress and how these effects vary by technology competence and technical support. In this sense, we found that TSCs increased burnout through techno-distress and job satisfaction through techno-eustress. The findings also revealed that TSCs’ indirect effects varied by technology competence. In conclusion, the research findings suggest that TSCs should not be viewed solely as a boon or a bane, offering a more nuanced perspective compared to previous studies. As seen in our study findings, TSCs’ contribution to a positive perception is likely to end up with a constructive (boon) outcome (e.g., job satisfaction). On the contrary, TSCs’ contribution to a negative perception is likely to result in a destructive (bane) consequence (e.g., burnout). Therefore, these findings highlight the importance of developing targeted strategies such as educational programs, workshops, and technology-oriented support interventions to create a perception of technostress that enhances job satisfaction and reduces burnout in nurses. Future research should thoroughly investigate both the positive and negative aspects of technostress across various healthcare settings and elaborate intervention approaches to increase its positive effects and reduce its negative consequences.

Abbreviations

TSCs

Technostress Creators

JD-R

Job Demands-Resources Model

Author contributions

KK: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing.MT: Conceptualization, Data curation, Investigation, Writing – original draft, Writing – review & editing.GA: Conceptualization, Writing – original draft, Writing – review & editing.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The datasets generated and analysed during the current study are not publicly available due to privacy and ethical restrictions of the participants, but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Nigde Omer Halisdemir University Ethic Committee granted formal approval for this study (E-86837521-050.99-299436). The research was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. All participants gave informed consent to participate prior to data collection.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.