Being healthy and achieving life harmony: the role of hybrid work and the mediating effect of work–family[with pets] conflict

Hybrid work has been increasingly adopted all over the world (Alexander et al., 2021). According to Beno (2021), the hybrid working model combines telework with in-person work from the office and is characterized by flexibility and less interference between work and family matters (Grant et al., 2019; Junça-Silva and Caetano, 2024). These characteristics are factors predicting employees’ well-being (Charalampous et al., 2023). Empirically, recent studies have indicated that hybrid work also allows greater flexibility and less work-family conflict (WFC; Choudhury et al., 2024; Naqshbandi et al., 2024).

WFC is a form of inter-role conflict that occurs when the energy, time, or behavioral demands of the work role conflicts with family life roles (Greenhaus and Beutell, 1985; Kossek and Lee, 2017). The role theory (Katz and Kahn, 1978) argues that role conflict occurs when individuals’ different roles, such as those about work and family, clash with one another (Kahn et al., 1964). Thus, based on the role theory, work-WFC arises when engagement in work-related activities interferes with concurrent family activities (Carlson et al., 2000), such as those involving family pets.

Despite the significant costs and time demands associated with pet ownership, the number of families owning pets continues to grow (Bussolari et al., 2021; Carroll et al., 2022). This trend highlights the increasing importance of investigating how human-animal relationships influence work-life balance, paralleling the well-documented effects of relationships with human family members. For instance, in 2022, Europe reported approximately 340 million companion animals, including 127 million cats, 104 million dogs, and substantial populations of birds, small mammals, fish, and reptiles, reflecting a rise of 27 million compared to 2021 (FEDIAF, 2023). Similarly, in Portugal, the Companion Animal Information System recorded approximately 3.1 million companion animals—comprising around 1.8 million cats, 2.6 million dogs, and smaller populations of birds, small mammals, and reptiles—an increase of 800,000 from the previous year (ICNF, 2022). These statistics underscore the growing relevance of understanding the role of pets in the work-life dynamics of modern families (Kogan et al., 2022).

Despite these trends, research remains limited on new forms of WFC arising from pet ownership. Addressing these forms of WFC within human-pet relationships is both timely and essential (Applebaum and Zsembik, 2020). Recently, Junça-Silva (2023a) proposed the concept of work-[pet]family conflict, a subtype of WFC. She suggested that, as pets are increasingly considered family members, work-[pet]family conflict arises when work demands interfere with responsibilities or activities related to pets (e.g. caregiving; Delanoeije and Verbruggen, 2024). Individuals may experience WFC when unable to fulfill pet-related obligations, similar to the conflicts experienced with human family members (Applebaum and Zsembik, 2020). Hence, similar to WFC, work-[pet]family conflict is likely to be related to various critical outcomes, such as well-being (Huo and Jiang, 2023) or mental health (Yuan et al., 2023).

Harmony is an indicator of well-being and includes an overall assessment of the extent to which each person feels balanced, integrated into their social environment, and in tune with their life (Kjell and Diener, 2021). Mental health is, according to the WHO (2023), a state of well-being in which individuals realize their own abilities, can cope with the normal stresses of life, can work productively, and can make a contribution to their community. WFC is a significant predictor of both well-being and mental health (Antino et al., 2022).

The Job Demands-Resources (JD-R) model (Demerouti et al., 2001; Bakker et al., 2023) provides valuable insights into how different work modalities, such as on-site, telework, and hybrid work, shape experiences of WFC and work-[pet]family conflict. By emphasizing the need for adequate job resources—such as flexible scheduling—the JD-R model highlights how work arrangements can be adapted to decrease the prevalence of WFC and work-[pet]family conflict. For instance, hybrid work models, which offer employees more control over their work environment and schedule, may serve as protective resources that reduce the intensity of these conflicts, promoting a healthier work-life balance and supporting mental health.

However, to date, there are no studies analyzing these variables together. Hence, regardless, of the type of WFC, and with the increasing adoption of the hybrid working model, it becomes relevant to understand whether this working model has an effect on WFC and work-[pet]family conflict and whether these impact well-being and mental health. Thus, intending to fill the gap on this topic, this study used the role theory to develop a conceptual model testing the indirect effect of the hybrid working model on well-being and mental health through decreases in WFC (Study 1) and work-[pet]family conflict (Study 2).

This study makes several theoretical contributions. First, it extends the JD-R model by investigating how hybrid work arrangements can mitigate specific types of role conflict (Grant et al., 2019; Charalampous et al., 2023). By examining the effects of hybrid work on both WFC and work-[pet]family conflict, this study highlights how flexible work arrangements can function as job resources (Bakker et al., 2023), reducing the psychological strain associated with these conflicts and contributing to enhanced well-being and mental health.

Additionally, this study expands the scope of role theory, which has traditionally focused on role conflicts involving human family members (Bakar and Salleh, 2015). By including role conflicts related to pet ownership responsibilities, this research broadens the application of role theory to capture the experiences of employees for whom pets are considered integral family members (Linne and Angilletta, 2024). This approach provides a more comprehensive understanding of how diverse role demands intersect and impact employees’ well-being and mental health (Applebaum and Zsembik, 2020), especially in modern contexts where pets play an increasingly prominent role in family life (Kogan et al., 2022).

Additionally, the study advocates for expanding work-life balance constructs to encompass not only family-related conflicts but also conflicts stemming from pet-related responsibilities. Hence, by conceptualizing and empirically testing work-[pet]family conflict as a distinct subtype of WFC, the study introduces a new dimension to the WFC literature (Junça-Silva, 2023a). Recognizing that pets are increasingly seen as family members (Barr et al., 2024), it highlights the unique demands pets can create and the resulting role conflicts (Allen, 2024). This adds depth to existing work-life conflict frameworks and integrates human-animal relationships into organizational behavior research which underscores the importance of recognizing diverse sources of conflict beyond traditional family relationships (Kogan et al., 2022). This broadened scope encourages further theoretical development to address the evolving nature of family structures and pet ownership in modern society.

As such, by exploring an unexplored construct as it is the work-[pet]family conflict, we answer the recent call for more studies (1) on the intersection of pets with work life (Kelemen et al., 2020) and (2) on the inclusion of pets as family members, that can give rise to role conflicts between their pet parents’ job activities and pet-related responsibilities (Junça-Silva, 2023a).

The study also contributes to theory by linking the hybrid work model to well-being and mental health, mediated by reductions in both traditional work-family conflict and work-[pet]family conflict. Demonstrating that hybrid work can mitigate these conflicts provides a theoretical basis for identifying specific work arrangements that may reduce stressors associated with competing role demands (Charalampous et al., 2023).

Practically it also has implications. First, since, in Portugal, this working model is relatively recent and lacks study (Junça-Silva, 2023b; Lopes et al., 2022), the current research becomes relevant to understanding the impact of this model on employees and organizations themselves. Therefore, by testing the mediating role of two forms of WFC, this study will enable organizations to understand the advantages and disadvantages of this working model and which strategies can be adopted to improve employees’ mental health. Lastly, this is relevant for practice as it can support managerial empirical-based decisions regarding strategies to promote their employees’ work-[pet]life balance.

Nilles (1998) first introduced the concept of telecommuting to define an agreement between employer and employee that allows work to be performed outside the usual workplace regularly, utilizing information and telecommunications technologies (ICT) to replicate significant aspects of the centralized work environment. Over time, the term telecommuting gradually evolved into teleworking, and recently other related concepts arose, such as home-based work, e-work, or remote work (Junça-Silva et al., 2022b).

Telework can be defined as work performed outside the conventional workplace (e.g. office), with communication done through ICT (Olson and Primps, 1984). Telework emerged as an organizational strategy for reducing costs (e.g. Egan, 1997) and addressing employees’ needs for work-family balance (e.g. Shamir and Salomon, 1985); however, telework remained relatively limited until the technological advancements and the recent changes in workplace dynamics (Hopkins and Bardoel, 2023).

Hybrid work has recently gained popularity and is defined as a work modality in which employees split their time between a traditional workplace and teleworking (usually from home or from “third places” like coworking spaces). This approach aims to combine the best aspects of both telework and office-based work (Naqshbandi et al., 2024). Introduced by Beno (2021), the hybrid model is characterized by its flexibility and the autonomy it offers employees in choosing between telework and in-office work (Choudhury et al., 2024; Sampat et al., 2022).

The transition to hybrid work has generated extensive debate, with research highlighting both its advantages and disadvantages (Almeida et al., 2024). On the one hand, some studies have raised concerns about potential downsides, such as diminished team cohesion, communication challenges, and difficulties in maintaining organizational culture (Charalampous et al., 2023; Wigert, 2022). Additionally, hybrid work may exacerbate inequalities, as not all employees have equal access to the resources or environments required for effective remote work (Koskela et al., 2020).

On the other hand, hybrid work arrangements have been shown to enhance employee well-being and job satisfaction by offering increased flexibility and autonomy (Bloom et al., 2023; Naqshbandi et al., 2024; Mutebi and Hobbs, 2022; Tsipursky, 2023). These arrangements enable greater control over the location and timing of work tasks, facilitating efficient time management and reducing interference between work and personal life (Grant et al., 2019; Wigert and Agrawal, 2022). Employees often report improved mental health, reduced stress from commuting, and better alignment of professional and personal responsibilities in hybrid models (Ergotron, 2022). Additionally, hybrid work has been linked to lower burnout rates, improved performance without adverse effects on career progression, and enhanced work experiences for employees (Almeida et al., 2024; Hopkins and Bardoel, 2023). Organizations also benefit, as hybrid work promotes productivity and reduces fixed costs (Naqshbandi et al., 2024), making it a potentially advantageous strategy for both employees and employers.

These mixed findings underscore the need for further research to identify the conditions under which hybrid work can maximize benefits while mitigating its potential drawbacks.

Hybrid work helps employees to better manage their work-life boundaries (Charalampous et al., 2023). Therefore, individuals who combine telework and in-person work are likely to reduce their WFC by improving their time management skills, facilitated by the flexibility that this arrangement offers (Junça-Silva, 2023b).

WFC can be understood as a specific subset of work-life conflict emerging when engagement in professional activities interferes with participation in family-related responsibilities or when occupational stress negatively impacts behavior within the family sphere (Greenhaus and Powell, 2003; Kossek and Lee, 2017). Thus, WFC encompasses a broad range of inter-role conflicts arising from diverse family-life demands, which may also include responsibilities toward pets as family members (Kossek and Lee, 2017).

Given the rising prevalence of pet ownership, especially among Millennials and Generation Z, pets (or companion animals) are increasingly considered integral family members (Barr et al., 2024; Dale, 2022). As with human family members, pet-related responsibilities and activities can conflict with work demands (Kogan et al., 2022), potentially leading to a distinct subtype of WFC—work-[pet]family conflict (Junça-Silva, 2023a).

“Work-[pet]family conflict occurs when work interferes with pet-family life or pet responsibilities.” (Junça-Silva, 2023a, p. 4). An example is when employees miss pet responsibilities (e.g. pet caregiving) because they have to be working at the office until late hours. Thus, either WFC or work-[pet]family conflict appear to be a form of inter-role conflict that is triggered by the incompatibility of different roles and arouses tension with [pet]family, or personal life goals (Kossek and Lee, 2017).

Greenhaus and Beutell (1985) identified three forms of WFC: (1) time-based conflict, (2) strain-based conflict, and (3) behavior-based conflict. Time-based conflict may occur when time devoted to one role hinders participation in another role (e.g. when one is working and cannot be available to take the pet to the veterinary or cannot pick up the child at school). Strain-based conflict suggests that strain experienced in one role interferes with participation in another role (e.g. when an argument with a spouse leaves a person in a bad mood at work, or when frustrations at work leave employees exhausted, decreasing their involvement with their pets at the end of the day). Finally, behavior-based conflict occurs when specific behaviors required in one role are incompatible with behavioral expectations in another role (e.g. treating a spouse like a coworker or treating the pet like a client; Greenhaus and Beutell, 1985). Regardless of the type of conflict, any one of them appears to be a significant predictor of well-being and mental health (French et al., 2018; Kossek and Lee, 2017).

Role theory (Katz and Kahn, 1978) provides a framework for understanding the origins and consequences of WFC and work-[pet]family conflict. According to role theory, individuals occupy multiple roles within both their work and personal lives, each with distinct expectations, responsibilities, and behavioral norms (Xu, 2009). Conflict arises when the demands of one role interfere with the ability to meet the expectations of another, leading to inter-role conflict (Greenhaus and Beutell, 1985). Thus, according to role theory WFC is a form of perceived role conflict, subjective in nature, and arises when individuals experience incompatibilities between their different roles (work and family life [with pets]) (Bakar and Salleh, 2015).

In the case of WFC, role theory suggests that professional obligations, such as long hours or job-related stress, can conflict with family-related responsibilities, thereby disrupting family involvement and impacting well-being (Kayaalp et al., 2021; Yuan et al., 2023). Similarly, work-[pet]family conflict occurs when the demands of work conflict with responsibilities associated with pet ownership (Hoffman, 2021; Kogan et al., 2022), a role increasingly viewed as akin to family in contemporary society (Martins et al., 2023). For individuals who consider pets as family members (Dale, 2022; Linne and Angilletta, 2024), responsibilities related to pet care and companionship may create role expectations that can clash with work demands (Delanoeije and Verbruggen, 2024), leading to unique forms of strain, such as feeling torn between work obligations and pet care needs (Applebaum and Zsembik, 2020). The consequences of these conflicts, as predicted by role theory, extend beyond immediate role strain, often impacting broader outcomes such as life harmony, mental health, and overall well-being (Delanoeije, 2020; Huo and Jiang, 2023; Junça Silva, 2024).

Moreover, by understanding how different work modalities (e.g. on-site, telework, hybrid work) influence these conflicts, JD-R model (Demerouti et al., 2001; Bakker et al., 2023) helps clarify why certain work arrangements may decrease or increase WFC and work-[pet]family conflict, thereby affecting employees’ well-being and mental health (Junça-Silva, 2023a, 2024).

Within the JD-R framework, WFC and work-[pet]family conflict function as significant job demands, imposing psychological strain on employees by requiring continuous effort to balance competing roles. When these conflicts are prolonged or unresolved, they may become chronic stressors, gradually depleting personal resources essential for maintaining mental health and well-being (Demerouti and Bakker, 2023). Over time, this depletion can lead to adverse outcomes, including increased exhaustion, diminished job satisfaction, and greater susceptibility to mental health impairment (Tran, 2023).

Applying the JD-R model to these conflicts also allows us to consider the role of job resources—such as flexibility in work arrangements—in mitigating the negative effects of WFC and work-[pet]family conflict (Costa et al., 2024; Huaman et al., 2023). For example, hybrid work arrangements, which allow employees to alternate between telework and on-site work (Grant et al., 2019) and offer them more control over their work environment and schedule, may serve as a valuable job resource by reducing the likelihood of work demands interfering with family responsibilities (Charalampous et al., 2023). This flexibility enables employees to better balance work and family (or pet) obligations, thus reducing the strain associated with these inter-role conflicts (Kogan et al., 2022). Thus, within the JD-R model, hybrid working models serve as resources that reduce the intensity of these conflicts, promoting a healthier work-life balance and supporting mental health and well-being.

Kjell et al. (2016) argued that harmony in life is a crucial indicator of a person’s well-being. Harmony was defined as “the perception of inner harmony, such as inner peace, self-acceptance, serenity, a sense of balance and fairness” (Delle et al., 2011, p. 199). Harmony in life includes a flexible global assessment of whether a person’s life is balanced, mindful, fits into their social environment, and is in tune with their life (Kjell and Diener, 2021).

In addition, mental health is essential for health and quality of life, because it is a resource for daily life, and contributes to the functioning of individuals, families, communities, and society (Barry, 2009). According to the WHO (2023), mental health is a state of well-being in which an individual realizes their own abilities, can cope with the normal stresses of life, can work productively, and is able to contribute to their community. This definition highlights various aspects of positive mental health, including subjective well-being and affective balance, and the development of capacities to manage life, maximize each individual’s potential, participate, and contribute to society.

Empirically, some studies have shown that hybrid work, by reducing interference between work and personal life, increases overall well-being (Grant et al., 2019; Junça-Silva et al., 2022a, b, c; Charalampous et al., 2023). Further, a Gallup study (2022) showed that the second reason people prefer hybrid work is the increase in overall well-being and mental health (Junça-Silva, 2023a, b) Wigert (2022). The Evolving Office stated that some of these reasons are that hybrid work allows employees to recover physical health, achieve a better balance between work and personal life, provide comfortable work environments and promote flexibility that increases job satisfaction and mental health (Ergotron, 2022). Moreover, 66% of hybrid workers stated that their mental health had improved because they had more time for themselves compared to full-time work (Choudhury et al., 2024). Other studies showed that hybrid work promoted higher levels of well-being and mental health because employees could use their spare time spending it with family and friends, exercising, or caregiving their pets during the day (Delanoeije and Verbruggen, 2024). In Portugal, some studies have evidenced that hybrid work, by promoting greater flexibility and less interference between work and personal life, not only reduces emotional exhaustion but also improves mental health (Junça Silva et al., 2022a, b; 2023a).

Based on role theory and the JD-R model, we propose that the work modality—whether fully on-site, teleworking, or hybrid—will impact life harmony and employees’ mental health by influencing both forms of WFC. Accordingly, the following hypotheses were formulated (see Figure 1).

This research includes two studies conducted at different times and with different samples. The first was a two-wave study with a one-week time lag and was conducted in 2023. This aimed to test the first hypotheses. In 2024, with a different sample, but the same two-wave design, another study was carried out to test hypotheses 2a and 2b.

The Ethics Committee of the author’s university approved the study before it started. Two waves of data were collected to minimize the potential common method variance problem. In the first stage (Time 1), 600 surveys were distributed to working adults. Participants were part of the researchers’ professional networks. This first contact was made by email in which the research purpose and scope were clarified, and the anonymity and confidentiality of the data were warranted to ensure that participation was voluntary. Those who answered this email received another one with the link for the Time 1 survey: this survey incorporated measures of work modality, WFC, and socio-demographic characteristics. Overall, 471 responses were received, yielding a response rate of 78.5%. At Time 2, one week later, questionnaires to measure harmony in life and mental health were sent to the 471 participants who answered the first survey. At this stage, 403 completed surveys were gathered, generating a response rate of 67.16%. However, only 376 valid responses were considered after excluding invalid surveys (completed in less than 2 min or perfunctory answers), with an overall response rate of 62.66%. According to a power analysis (effect size of 0.2, error probability of 0.05), this sample size was considered sufficient.

Data was collected between October to November 2023. In addition to the two-wave data collection, other precautionary measures were used to minimize potential common method bias (CMB) (Podsakoff et al., 2024). All participants voluntarily and anonymously answered the online survey. Further, the items were randomized, and attention was set to screening questions in both surveys.

The participants included working adults from managerial positions working in Portugal. Of the overall sample, 60% were female. The mean age was 35 years (SD = 12.10), and the mean organizational tenure was 8 years (SD = 9.67). Participants reported working on average 36 h per week (SD = 11.71). In terms of the work arrangement carried out in the last week, 70% were engaged in hybrid work, 25% had performed fully on-site work, and only 5% had worked entirely remotely.

All the scales used have shown high reliability and validity in previous research. All the variables were measured with a five-point Likert scale. Following Brislin’s (1986) cross-cultural translation procedure, two bilingual academic researchers conducted the back-translation process to translate the English items into Portuguese. Moreover, we invited two experts in well-being studies to review the initial draft of the Portuguese surveys and to make some revisions to the item wording and instructions for respondents to increase the content validity (Wang et al., 2021). A pilot study with 55 employees in Portuguese organizations (not part of the final sample) was initially conducted to assure content validity.

The variable “work modality” was measured using the question “During the last week, what type of work arrangement were you in?”, with response options being: (1) Fully On-Site Work; (2) Fully Telework; and (3) Hybrid Work.

WFC was measured using the abbreviated 3-item version of Carlson et al’s. (2000) multidimensional measure of WFC (Matthews et al., 2010). The scale measured the three dimensions of WFC: time-based: “I have to miss family activities because of the amount of time I need to devote to work responsibilities.”; tension-based: “I often feel so emotionally drained when I come home from work that it prevents me from contributing to my family.”; and behavior-based: “The behaviors I engage in that make me effective at work do not help me to be a better parent and spouse.” These items were answered on a Likert scale ranging from 1 to 5 (Strongly Disagree to Strongly Agree). The Cronbach’s α in this study was 0.83 and the McDonald’s was 0.84.

Harmony was measured using the abbreviated three-item version of the Harmony in Life Scale (Kjell and Diener, 2021). An example item was “My lifestyle allowed me to be in harmony.” Participants responded on a 5-point Likert scale (1 - Strongly Disagree; 5 - Strongly Agree). The Cronbach’s α in this study was 0.91 and the McDonald’s was 0.92.

Mental health was measured using three items from the SF-36v2 Vitality and Mental Health Scales (e.g. “How often have you felt calm and relaxed in the past week?”) developed by Ware et al. (1996). Participants responded on a Likert scale from 1 (Never) to 5 (Always). The Cronbach’s alpha was 0.70 and the McDonald’s was 0.72.

Sex and age of the participants were used as control variables. Sex was used as a control because some studies have shown that women tend to be happier than men (Diener et al., 2020); therefore, differences between men and women could influence the outcome variables (i.e. harmony and mental health). Additionally, age may also account for influences on work-family conflict and mental health, as differences have been identified in how older and younger individuals experience professional life and their levels of mental health (Livingstone and Isaacowitz, 2018).

In the proposed mediating model (see Figure 1), there were three types of variables: (1) predictor (work modality); (2) two criterion variables (i.e. harmony and mental health); and (3) one mediator (WFC). SPSS 28.0 and the software JASP (version 0.14.1) were used to test the proposed research models. First, descriptive analysis was conducted to calculate the mean and standard deviation for each variable. Second, correlational analyses were performed to examine whether work modality was associated with the mediator and the criterion variables. Third, the measurement model’s goodness of fit was evaluated. In this regard, we found that the Root Mean Square Error of Approximation (RMSEA) < 0.08, Standardized Root Mean Squared Residual (SRMR) < 0.08, Comparative Fit Index (CFI) > 0.90, and Tucker-Lewis Index (TLI) > 0.90 evidenced a good fit (Kline, 2015).

Although we have followed some recommended procedures to reduce the potential common method bias–i.e. using closed-ended questions mixed in the survey (e.g. “I like pets”) and resorting to previously validated surveys to assess the variables under study–it cannot be completely avoided (Podsakoff et al., 2024). Hence, to understand its presence in the study we followed some recommendations.

First, we performed Harman’s single factor test to check for common method bias. The findings showed that the first factor only accounted for 40.83% of the total explained variance; hence, the common method bias was not a serious issue.

Second, as Kock (2015) suggested, we also performed a full collinearity evaluation test to check for the potential common method bias. The results demonstrated that all the variance inflation factor values ranged from 1.02 to 1.48; because the values were less than the cut-off point of 3.33, multicollinearity concern was not a severe issue in this study.

At last, we performed three confirmatory factor analyses (CFA) to test the independence of the variables under study. To assess the adequacy of the model and compare it with other reasonable alternative models, we analyzed diverse fit indices, namely CFI, TLI, SRMR, and RMSEA (Hair et al., 2014). Model 1 was the hypothetical three-factor model comprising separate scales for WFC, harmony, and mental health. Model 2 was a two-factor model where harmony and mental health were combined into a single factor, along with WFC loaded onto another factor. Model 3 was a single-factor solution where all items were loaded onto a single latent factor. Table 1 shows that the three-factor model (Model 1) provided the best fit to the data (χ²/df = 1.49, p < 0.001, CFI = 0.99, TLI = 0.99, SRMR = 0.05, and RMSEA = 0.04 CI 95% [0.01, 0.07]), and all other alternative models showed poorer fit. These results, along with reliability indices measured through Cronbach’s alpha in all measurement scales, demonstrated the discriminant and convergent validity of the study; therefore, we proceeded with testing the two hypotheses.

Table 2 shows the correlations between the variables, as well as their mean values and standard deviations. According to Field (2009), relatively small standard deviations compared to the means of the variables suggested that the means represented the observed data. The results also showed that all variables were significantly correlated with each other, in the expected direction.

As observed in Table 2, the reliability of the study variables was above the recommended threshold of 0.70, in line with Fornell and Larcker (1981). The result of convergent validity, which measures how the indicators of the latent construct correlate, revealed that the values of Average Variance Extracted (AVE) for all latent constructs in the study were above 0.5. Additionally, the AVE for each construct was evaluated concerning its correlation with other constructs, and the AVE value was found to be higher than the correlation of the construct with other constructs, thus supporting convergent validity. The discriminant validity demonstrates how the indicators of each latent variable are unique, and the square roots of the Average Variance Extracted (AVE) indicated by the diagonal value of each latent variable were all greater than the correlations of each variable. Maximum Shared Variance (MSV) was also analyzed; the results of MSV showed that it was lower than AVE for all constructs; thus, discriminant validity was supported. In this way, the reliability, convergent validity, and discriminant validity of the study were confirmed. Based on the validity of the study instrument, the study hypotheses were analyzed.

The structural equation model fit the data well: χ²_(df) = 2.18, p < 0.001, CFI = 0.97, TLI = 0.95, RMSEA = 0.06, 90% CI [0.00; 0.13]), SRMR = 0.05. The standardized path coefficients between the variables are presented in Figure 2.

Firstly, the test of the indirect effect showed that WFC significantly mediated the relationship between the work regime and harmony (β = 0.10; p < 0.05; 95% CI [0.00; 0.20]). The model explained 11% of the variance in harmony (R² = 0.11). Thus, hypothesis 1a was supported by the data.

Secondly, the results showed a similar pattern for mental health, i.e. the coefficient associated with the indirect effect was statistically significant (β = 0.11; p < 0.05; 95% CI [0.00; 0.21]). Overall, the model explained 13% of the variance in mental health (R² = 0.13) (Table 3); therefore, hypothesis 1b also received support from the data.

This study shows that WFC is an explanatory mechanism of the relationship between work modality and well-being indicators (i.e. harmony and mental health). Specifically, it shows that those who are in the hybrid work model tend to have lower levels of WFC that, in turn, appears to promote their life harmony and mental health.

The second study will test the same model but with work-[pet]family conflict.

We followed the same procedures as the first study., however we added some selection criteria. The selection criteria for this study focused on individuals from Generation Y (Millennials; born between 1980 and 1996) and Generation Z (born between 1997 and 2012) (Mahmoud et al., 2021). Additionally, to ensure relevant insights into workplace dynamics and pet ownership, only participants with pets and a minimum of one year of professional experience were included. This criterion was chosen to capture individuals who have had sufficient exposure to workplace environments, allowing them to provide informed perspectives on work-related conflicts. By focusing on these generational cohorts, the study aimed to explore the unique experiences and challenges faced by Millennials and Generation Z in balancing professional responsibilities with personal and family-related obligations, including those related to pet ownership (Dale, 2022).

Similar to Study 1, the Ethics Committee approved this two-wave study. At Time 1, 1021 surveys were distributed by email to working adults who were part of the researchers’ professional networks. This email asked for their collaboration, explained the study’s goals, and ensured the anonymity and confidentiality of the data. It was also emphasized that their participation was voluntary and that they could withdraw at any moment. Those who answered this email received another one with the link for the Time 1 survey that assessed work modality, work-[pet]family conflict, and socio-demographic characteristics. Overall, 778 responses were received, yielding a response rate of 76.19%. At Time 2, one week later, surveys measuring life harmony and mental health were sent to the 778 participants. At this stage, 589 completed surveys were gathered, generating a response rate of 57.68%. However, only 479 valid responses were considered after excluding invalid surveys (completed in less than 2 min or perfunctory answers or mentioning not having pets – as this was a criterion for this study), with an overall response rate of 46.91%. According to a power analysis (effect size of 0.2, error probability of 0.05), this sample size was considered sufficient. Data was collected between January to March 2024.

The participants included working adults from administrative (46%) and managerial positions (54%) working in Portugal. Of the overall sample, 67.5% were female. The mean age was 33.39 years (SD = 13.76), and the mean organizational tenure was 7.43 years (SD = 5.7). Approximately 45.4% of the participants belonged to Generation Z, while the remaining participants were from Generation Y (Millennials). In terms of the work modality carried out in the last week, 77% were engaged in hybrid work, 10% had performed fully on-site work, and 13% had worked entirely remotely. All participants were pet owners, and the mean number of pets was 1.35 (SD = 1.41), of which 94% reported owning dogs and 44% cats.

All the scales used have shown high reliability and validity in previous research.

We used the same initial question as we used in the first study.

We used the work-[pet]family boundaries scale (Junça-Silva, 2023a, b). Three items were used to assess time-based conflict (“I have to miss activities with my pets (or engage in fewer activities with them) due to the amount of time I must spend on work responsibilities.”), tension-based conflict (“I am often so emotionally drained when I get home from work that it prevents me from contributing to my pets.”), and behavioral-based conflict (“The behaviors I perform that make me effective at work do not help me to be a better pet parent.”). The items were answered on a 5-point Likert scale (1 - Strongly Disagree; 5 - Strongly Agree) (α = 0.90; ω = 0.90).

We used the same measure of Study 1 (i.e. the abbreviated three-item version of the Harmony in Life Scale; Kjell and Diener, 2021; α = 0.91; ω = 0.91).

We used the same measure of Study 1 (Ware et al., 1996; (α = 0.76; ω = 0.75).

The sex and age of the participants were used as control variables.

The same procedures were followed, and the model was tested on JASP software.

First, we performed Harman’s single-factor test to check for common method bias. The findings showed that the first factor only accounted for 42.52% of the total explained variance; hence, the common method bias was not a serious issue.

Second, as Kock (2015) suggested, we also performed a full collinearity evaluation test to check for the potential common method bias. The results demonstrated that all the variance inflation factor values ranged from 1.03 to 1.10; because the values were less than the cut-off point of 3.33, multicollinearity concern was not a severe issue in this study.

At last, we performed three confirmatory factor analyses (CFA). Model 1 was the hypothetical three-factor model comprising separate scales for work-[pet] family conflict, harmony, and mental health. Model 2 was a two-factor model where harmony and mental health were combined into a single factor. Model 3 was a single-factor solution where all items were loaded onto a single latent factor. Table 4 shows that the three-factor model (Model 1) provided the best fit to the data (χ²/df = 2.08, p < 0.001, CFI = 0.99, TLI = 0.99, SRMR = 0.03, and RMSEA = 0.04 CI 95% [0.02, 0.05]), and all other alternative models showed poorer fit. These results showed the discriminant and convergent validity of the study; therefore, we proceeded with testing H2a and H2b.

Table 5 shows the correlations between the variables, as well as their mean values and standard deviations.

The result of convergent validity revealed that the values of Average Variance Extracted (AVE) for all latent constructs in the study were above 0.5. The square roots of the Average Variance Extracted (AVE) were all greater than the correlations of each variable. The results of the MSV showed that it was lower than AVE for all constructs; thus, discriminant validity was supported. Hence, the reliability, convergent, and discriminant validity of the study were confirmed. Thus, the hypotheses were tested.

The structural equation model fit the data well: χ²_(df) = 2.02, p < 0.001, CFI = 0.99, TLI = 0.99, RMSEA = 0.04, SRMR = 0.03. The standardized path coefficients between the variables are presented in Figure 3.

Firstly, the test of the indirect effect showed that work-[pet]family conflict significantly mediated the relationship between the work modality and harmony (β = 0.05; p < 0.05; 95% CI [0.01; 0.10]). The model explained 14% of the variance in harmony (R² = 0.143). Thus, hypothesis 2a was supported by the data.

Secondly, the results showed a similar pattern for mental health, i.e. the coefficient associated with the indirect effect was statistically significant (β = 0.04; p < 0.05; 95% CI [0.01; 0.08]). Overall, the model explained 7% of the variance in mental health (R² = 0.074) (Table 3); therefore, hypothesis 2b also received support from the data (see Table 6).

This research relies on role theory and JD-R to design a conceptual model testing whether WFC and work-[pet]family conflict mediates the relationship between the modality of work (fully on-site, versus fully telework versus hybrid work) and harmony and mental health.

This study is particularly timely given the widespread adoption of hybrid work arrangements (Hodzic et al., 2024) and the evolving composition of modern families, who increasingly regard pets as family members (Linne and Angilletta, 2024). Unlike previous generations, today’s families, especially Millennials and Generation Z, often treat their pets as integral members of the household (Dale, 2022), and they tend to feel a strong sense of responsibility and even guilt when work obligations prevent them from attending to their pets’ needs (Kogan et al., 2022). These modern attitudes toward pet ownership mirror traditional family dynamics, where leaving a family member unattended or missing responsibilities can lead to stress and guilt (Foucreault et al., 2023). With hybrid work arrangements becoming a permanent feature in many organizations, understanding how this work modality affects both human and pet-family conflicts is crucial (Junça Silva, 2024). This research sheds light on the unique pressures faced by pet-owning employees in a hybrid work environment, emphasizing the need to address work-family and work-[pet]family conflicts to support employee well-being more comprehensively.

Overall, both studies appear to suggest that both types of inter-role conflict are mechanisms through which work modality, particularly hybrid work, influence both well-being indicators (i.e. harmony and mental health).

Hybrid work, by offering enhanced flexibility and control over tasks, schedules, and time allocation (Grant et al., 2019), creates favorable conditions for employees to better manage the demands of both work and family life (Almeida et al., 2024). This flexibility is crucial in reducing perceived WFC and work-[pet]family conflict, as it allows employees to respond more effectively to family responsibilities without compromising professional obligations. The reduction in these conflicts helps alleviate feelings of guilt or tension that often arise from an inability to balance both domains (Aarntzen et al., 2023; Foucreault et al., 2023), leading to improved mental health and greater life harmony (Korabik, 2017). By enabling a more seamless integration of work and personal responsibilities, hybrid work fosters a supportive environment that minimizes the strain of inter-role conflicts, ultimately promoting well-being and satisfaction across both personal and professional spheres.

This research makes significant contributions to the literature on work-life boundaries by addressing the complex, concurrent roles that individuals manage, often leading to inter-role conflicts such as WFC and work-[pet]family conflict (Kossek and Lee, 2017). These conflicts are critical to examine given their potential negative effects on well-being and mental health (Huaman et al., 2023; Kayaalp et al., 2021).

This study also expands the application of role theory, which has traditionally examined role conflicts in the context of human family members (Bakar and Salleh, 2015). By considering role conflicts related to pet-care responsibilities, this research broadens role theory’s scope, acknowledging that for many employees, pets are regarded as integral family members. This expanded approach enables a more nuanced understanding of how diverse role demands intersect and influence well-being and mental health (Applebaum and Zsembik, 2020), particularly in contemporary contexts where pets occupy an increasingly significant place within family life (Kogan et al., 2022).

Indeed, a major contribution of this study is its investigation of an overlooked subset of WFC—the intersection between work demands and pet-care responsibilities, termed work-[pet]family conflict (Junça-Silva, 2023a). By examining this specific conflict, the study addresses calls for greater understanding of how pets are integrated into work-life dynamics (Kelemen et al., 2020) and opens new avenues for investigating the mental health impacts on pet-owning employees who strive to balance work and pet-family obligations (Junça-Silva, 2023a; Kogan et al., 2022).

Lastly, this research extends the JD-R model by investigating how hybrid work arrangements can diminish both types of inter-role conflict (Grant et al., 2019; Charalampous et al., 2023). By examining the effects of hybrid work on both WFC and work-[pet]family conflict, this study highlights how flexible work arrangements can function as job resources (Hodzic et al., 2024), reducing the psychological strain associated with these conflicts and contributing to enhanced well-being and mental health.

The results show that WFC mediates the relationship between work modality and harmony and mental health. In other words, employees in a hybrid work regime tend to have lower levels of WFC, which in turn tends to increase their life harmony and mental health states. Empirical studies have shown that the shift to hybrid work offers multiple benefits for employees (e.g. Andrade and Petiz Lousã, 2021; Choudhury et al., 2024). Beyond enhancing flexibility and facilitating a balance between professional and personal life (Charalampous et al., 2019), hybrid work arrangements are associated with increased well-being and happiness (Naqshbandi et al., 2024). Consequently, it is unsurprising that employees express a preference for hybrid work (Wigert, 2022), as this model enables them to adopt healthier lifestyles and reduce psychological strain, such as stress from commuting. Hybrid work also promotes work-life balance and provides flexibility, which supports better mental health (Ergotron, 2022). In the 2022 IWG study, 66% of hybrid workers reported improved mental health due to hybrid work, attributing this improvement to increased time for personal and family activities (e.g. spending time with loved ones, exercising, or caring for pets; Tsipursky, 2023) and to a reduction in WFC (Junça-Silva, 2023b). Thus, hybrid work not only reduces WFC but also fosters greater life harmony and mental well-being for employees.

Work-[pet]family conflict also appears to be mitigated in hybrid work models, aligning with increases in employees’ life harmony and mental health. Given that younger generations, such as Millennials and Generation Z, tend to form strong attachments to their pets (Dale, 2022) and are more likely to have pets in their households, it is increasingly relevant to consider pets as integral family members (Linne and Angilletta, 2024). This shift in perspective necessitates an awareness of potential conflicts that arise not only from pet-care responsibilities, such as veterinary visits, but also from pets’ social needs, like companionship, amidst the demands of work and family life (Kogan et al., 2022). These pet-related conflicts are comparable to traditional work-family conflict (Huo and Jiang, 2023), with notable implications for employees’ well-being and mental health (Delanoeije, 2020; Junça-Silva, 2023a).

This research demonstrates that hybrid work arrangements offer the flexibility and autonomy essential for pet-owning families, supporting a healthier balance between work and pet-care responsibilities (Applebaum and Zsembik, 2020). By granting employees greater control over when and where they perform work tasks, hybrid work models contribute to improved mental health and well-being (Hopkins and Bardoel, 2023). In summary, hybrid work reduces both WFC and work-[pet]family conflict, thereby enhancing employees’ life harmony and mental health.

The study highlights the positive impact of hybrid work arrangements on employees’ well-being, particularly in the context of Millennials and Generation Z, who value flexibility and work-life balance. By adopting hybrid work models, organizations can foster improved mental health, reduce stress, and enhance life harmony among these generations. Further, Millennials and Generation Z, who are more likely to be pet owners (Junça-Silva, 2023a), prioritize workplaces that offer flexibility to balance work and family (including pet-related responsibilities). Organizations that offer hybrid work options can attract and retain this talent pool by addressing their need for flexibility and accommodating their desire to maintain a balance between their professional and personal lives. Plus, given the growing trend of pet ownership among Millennials and Generation Z, the study emphasizes the importance of considering pets as integral family members. Hybrid work models, by providing flexibility, allow employees to better manage work and pet-care responsibilities, reducing work-[pet]family conflicts and improving overall well-being.

In sum, this study suggests that organizations that implement hybrid work models are not only improving employee well-being and mental health but also positioning themselves as attractive employers for Millennials and Generation Z, who increasingly view their pets as family members and value a flexible work-life integration.

This research has some limitations. Firstly, the studies used self-reported measures, i.e. measures based on an individual’s report of their symptoms, behaviors, beliefs, or attitudes (Levin-Aspenson and Watson, 2018). This type of data collection has some limitations, such as individuals being often biased when reporting their own experiences–social desirability (Devaux and Sassi, 2015), which may limit the reliability of the results obtained.

A limitation of this study is the overly simplistic measurement of the hybrid work modality variable. This approach raises concerns regarding its accuracy and the extent to which it captures the complexities of hybrid work arrangements and their influence on the variables under investigation. As a result, the current measurement may lack validity and comparability with other study variables, potentially compromising the robustness and generalizability of the findings. Future research would benefit from adopting a more nuanced approach to measuring hybrid work, such as assessing the frequency and distribution of telework versus on-site workdays, to provide a deeper understanding of the diverse experiences and impacts of hybrid work arrangements.

Finally, the study’s two-wave design, i.e. the data is analyzed at two time points, is also a limitation. Although this type of design has advantages such as the quick collection of data and the ability to collect it on different variables to see how they influence a certain condition, it also has some disadvantages (Cherry, 2022). It is not always possible to be sure that the conditions that this type of study measures are the result of the influence of a certain factor. In many cases, differences between individuals can be attributed to variation among the study’s participants. Thus, cause-effect relationships are more difficult to determine than, for example, in a longitudinal study (Cherry, 2022). Additionally, these studies can lead to common method bias. Nevertheless, some measures were taken to identify the presence of common method bias, including confirmatory factor analyses and the test of the measures’ validity and reliability. These analyses show that the presence of common method bias is not a significant problem in the data.

Given this, it is suggested for future research to use a larger sample to draw more reliable conclusions. Future studies could also rely on longitudinal or daily designs, i.e. over several days, to analyze fluctuations in employees’ well-being and mental health. Furthermore, future studies should delve deeper into the examination of contemporary manifestations of WFC, such as the work-[pet]family conflict. An area ripe for future research could involve investigating the dynamics of work-[pet]family conflict, including its timing and mechanisms, and its intersections with pertinent work-related behaviors, such as performance, and their implications for overall well-being.

This study is significant in demonstrating the positive impact of hybrid work on overall well-being, providing employees with a flexible approach to balancing professional commitments with family life, including responsibilities related to pet ownership. Specifically, the findings highlight how hybrid work arrangements reduce levels of work-family conflict (WFC) and work-[pet]family conflict, fostering greater harmony and improved mental health across both work and family domains. These insights suggest that organizations can leverage the flexibility of hybrid work to enhance employee well-being, promoting a workplace culture that values balance and mental wellness.

I would like to thank my loyal companion, Kitty, whose silent presence, unconditional love and understanding gaze made every moment of writing lighter. This work is also yours – for the hours spent by my side, warming my days and reminding me of the true meaning of companionship and devotion.

Funding: This work was supported by Fundação para a Ciência e a Tecnologia, grant UIDB/00315/2020.