Examining Employee Well-Being in UAE Organizations: An Analysis within the Job Demands–Resources (JD-R) Framework

Stress and burnout: unaddressed mental health challenges among employees in the UAE

Research from the United Arab Emirates (UAE) indicates elevated yet inconsistent levels of stress and burnout across the healthcare, hospitality, and public sectors. A significant body of evidence highlights that these outcomes are primarily driven by demanding occupational factors, including high workloads, shift work, role ambiguity, poor work–life balance, and insufficient organizational support.^11–15 Despite the introduction of wellness programs, these core issues are often only partially addressed. Furthermore, the broader work environment specifically an institution’s organizational culture and prevailing leadership styles serves as a strong predictor of employee stress levels and overall well-being.¹¹ In Sharjah’s public sector, employees typically report considerable overall job satisfaction; however, moderate-to-high levels of depersonalization and diminished personal accomplishment are prevalent and significantly correlated with reduced happiness.¹⁶ In UAE hotels, burnout markedly increases psychological discomfort and the propensity to resign, with distress serving as a principal mechanism in explaining the relationship between burnout and staff turnover.¹⁷ Table 1 below shows the sector-wise well-being issues in the UAE.

The struggle of “One-Size-Fits-All” well-being programs in the UAE

Although workplace well-being programs are becoming increasingly common, empirical evidence concerning their effectiveness is notably inconsistent, especially when interventions adopt a generic, individual-centered approach instead of being customized to particular business contexts and workforce requirements. Numerous umbrella and scoping reviews demonstrate that universal psychological treatments have only small to moderate effects, often supported by low to very low-certainty evidence.^20–23 A review of 75 eHealth experiments indicated that average effects on depression, anxiety, and stress are minimal and have remained static over the past decade.²⁴ Moreover, universal eHealth stress management interventions implemented among unselected employees have shown no significant positive outcomes and, in some cases, have indicated potential harm, whereas the same content was effective when specifically targeted at highly stressed workers.²⁵ This deficiency is further substantiated by a comprehensive UK survey involving 46,336 workers, which found that participants in individual-level interventions did not exhibit superior outcomes compared to non-participants.²⁶ The UAE workforce exhibits a distinctive demographic and epidemiological profile; 89% of the population comprises expatriates from over 200 countries and faces a significant prevalence of chronic illnesses such as obesity, diabetes, and hypertension.²⁷

Policy vs. practice gap: disconnect between policies and the actual situation

Despite governmental initiatives such as the Ministry of Happiness and the National Program for Happiness aimed at prioritizing wellbeing in policy,^6,28,29 significant implementation gaps endure, resulting in disparate outcomes, especially for marginalized populations, including low-status expatriates and female administrators.^30–32 The COVID-19 pandemic highlighted disparities in mental health and resilience among employees.^12,33–35 Migrant workers face heightened risks of depression (25%) and suicide ideation (6.3%), with factors such as poor income and extended working hours exacerbating their vulnerability.³⁶ Table 2 below presents the current levels of workplace stress and the prevalence of health issues in the UAE.

Lack of measured impact

Research on employee well-being in the UAE is expanding, with numerous studies assessing its impact on outcomes such as performance, innovation, job satisfaction, and health. Yet coverage remains fragmented by sector and outcome, and measurement tools are not yet standardized. A global review notes that many worker wellbeing instruments lack full psychometric robustness and cross-study standardization.^38,39 Rigorous long-term and economic impact evaluations within the UAE remain limited.^40,41

Research gap

Although a growing body of literature documents employee well-being challenges in the UAE, significant gaps persist. Most studies focus narrowly on a single sector or a limited set of outcomes; cross-sector, multi-construct analyses are rare. More critically, despite the theoretical prominence of the JD-R model globally, no empirical study has systematically examined the mediating role of job crafting as a bridge between organizational/personal resources and performance outcomes within a UAE context using PLS-SEM. Furthermore, while prior UAE research documents burnout and exhaustion, the downstream behavioral mechanisms linking these states to job performance through proactive employee strategies remain theoretically underexplored in this setting. The present study addresses these gaps by applying an expanded JD-R framework to a multi-sector UAE sample, treating job crafting, work engagement, and exhaustion as simultaneous mediators in a structural model, thereby providing the first PLS-SEM-based test of the full JD-R motivational and health-impairment pathways in the UAE organizational context.

High stress and burnout: prevalence in the UAE

The welfare of employees in the United Arab Emirates (UAE) is shaped by a distinctive demographic profile, rapid economic development, and national policies that emphasize happiness and quality of life. Yet challenges persist, including high rates of chronic diseases, occupational stress, burnout, and mental health disorders across sectors such as healthcare, education, construction, and among migrant workers.^{16,27,30,31,33,36,42,43} Chronic disease risk factors are highly prevalent among UAE employees and account for almost 65% of national mortality.²⁷ Beyond physical health, burnout and emotional exhaustion negatively affect job satisfaction. Substantial cross-sectional evidence links burnout with reduced satisfaction and performance.^16,37 Leadership and organizational support emerge as crucial for improving workplace happiness; multiple studies show that top management commitment enhances retention and performance.^1,6,44 Gender and nationality disparities persist in economic perceptions and mental health outcomes.⁴³ Migrant and low-status blue-collar workers face even greater mental health risks, with depression and suicidal ideation more prevalent among these laborers.³⁶ National happiness initiatives appear to have limited micro-level impact, with persistent gaps between macro policy visions and individual employee experiences.^28,29,45,46

Consequences and Under-Addressed Challenges

Burnout is associated with significant repercussions globally, such as physical ailments, depression, insomnia, injuries, and elevated levels of absenteeism and presenteeism.^47,48 The Gulf Cooperation Council (GCC) region closely mirrors this global trend, as healthcare personnel exhibited elevated levels of stress, anxiety, and burnout during the COVID-19 pandemic.^49,50 The effect is also apparent in Dubai government institutions, where unaddressed mental health issues correlate with significant presenteeism, 20% absenteeism, and a 15% increase in staff turnover. Nonetheless, tackling these challenges is often hindered by cultural stigma; in the UAE, stigma leads 60% of affected employees to deliberately refrain from seeking assistance.¹⁹ The establishment of formal mental health programs within Dubai government entities is associated with a 30% increase in employee satisfaction and an estimated 4:1 return on investment.¹⁹ Table 3 presents the Evidence on UAE work stress.

Moving toward tailored frameworks

In light of the shortcomings of generic methodologies, contemporary conceptual and review literature strongly supports a paradigm shift away from static, top-down solutions. Scholars advocate for proactive, customized, and multifaceted frameworks that are intricately aligned with comprehensive organizational policies.^41,53–55 In uniquely diversified and high-risk situations such as the UAE, implementing these principles requires a thoughtful, context-specific approach. The design of well-being initiatives must be based on local empirical data on specific health risks and workforce demographics.²⁷ Successful programs cannot depend solely on individual coping strategies; instead, they must amalgamate individual-level instruments such as mindfulness, positive psychology, and resilience training with structural organizational adjustments, including job redesign, flexible scheduling, and participatory work practices.^53,55–57

The imperative for systemic change over individual-level fixes

A critical structural deficiency in several modern well-being programs is their excessive focus on individual behavior modification, neglecting the broader organizational elements. Numerous extensive analyses indicate that conventional wellness programs often overlook essential systemic factors influencing occupational health, including workload, scheduling, job design, and the prevailing organizational culture.^41,53–55 Individual-level interventions such as resilience training or mindfulness programs frequently fail to yield significant results due to inadequate implementation and a fundamental disconnect from essential structural changes.^26,54 Contrary to shallow add-on health initiatives, evidence increasingly substantiates the effectiveness of systemic, organizational-level transformations. Interventions that fundamentally transform working conditions such as extensive job and task redesign, the implementation of flexible working arrangements, and participatory organizational modifications exhibit markedly more consistent and enduring improvements in employee well-being.^56,57

Core pathways: demands vs. resources

The Job Demands–Resources (JD-R) model explains work in terms of two interrelated dimensions: demands that deplete energy and resources that enhance motivation. These two aspects influence two trajectories that determine burnout, engagement, health, and performance.^58–60 Job demands encompass elements of work that necessitate prolonged physical, cognitive, or emotional exertion and are associated with costs such as workload, time constraints, emotional pressures, and work–family conflict.^58,61,62 Elevated or persistent expectations can diminish well-being, heighten stress, and increase the risk of burnout. Conversely, job resources encompass physical, psychological, social, or organizational attributes that facilitate goal attainment, alleviate demands, and foster development (such as autonomy, feedback, social support, and equitable leadership). In summary, demands deplete energy, whereas resources cultivate motivation and resilience.

Interactions and performance outcomes

The model Figure 1 provides insight into how workplace stress and support converge to affect fatigue and engagement, thereby impacting performance and other outcomes. Core processes differentiate between the health-impairment trajectory, wherein job demands deplete energy, leading to exhaustion, burnout, and deteriorating health^{58,60,61,63,64} and a motivational trajectory, where job resources like autonomy, feedback, and supportive development enhance engagement and well-being.^{58,59,61,63,65,66} Exhaustion and burnout correlate with increased turnover intentions and diminished satisfaction and service quality, while engagement is positively connected with satisfaction and perceived performance.^61,66,67

Figure 1. The Job Demands-Resources (JD-R) Model (Bakker & Demerouti, 2017).⁸⁵

The model illustrates the dual-process framework comprising the motivational pathway (Job Resources → Job Crafting → Work Engagement → Job Performance) and the health-impairment pathway (Job Demands → Exhaustion → Job Performance). Personal resources are depicted as moderating both pathways. The expanded model includes job crafting as a proactive behavioral mechanism mediating the resource–performance relationship.

Dynamic feedback loops: job crafting

The JD-R framework views job crafting as an active, ongoing process in which employees continually modify their work to fit their needs and strengths, creating feedback loops. When crafted well, they boost engagement and access to resources, fueling an upward spiral of even greater engagement. However, if coping strategies are maladaptive and demands exceed capacity, stress builds, self-regulation falters, and burnout can lead to a downward spiral of strain and fatigue.^58,68,69 Job crafting renders the JD-R model dynamic: engaged and resourced employees are inclined to seek additional resources and meaning, whereas prolonged high demands may result in maladaptive cycles and burnout.

Research hypotheses

Based on the JD-R theoretical framework and the reviewed literature, the following hypotheses are formally stated:

Research objectives

This study pursues three research objectives:

Ethical approval and informed consent

This study received ethical approval from the Institutional Review Board (IRB) of Abu Dhabi University prior to commencement of data collection (File Number: CAESS-7; IRB Chair: Rania Al Dweik; Approval Date: 9 November 2025; Expiry Date: 9 November 2026). All procedures were conducted in accordance with the Declaration of Helsinki and applicable national guidelines for research involving human participants. Prior to participation, all respondents were provided with a detailed study information sheet outlining the purpose of the study, data confidentiality measures, and their right to withdraw at any time without consequence. Written informed consent was obtained from all participants before the questionnaire was administered. Responses were fully anonymized during analysis to ensure participant confidentiality. The IRB approval certificate is submitted as a supplementary document alongside this manuscript.

Research design and justification for convenience sampling

This study adopts a quantitative, cross-sectional research design employing Partial Least Squares Structural Equation Modelling (PLS-SEM). PLS-SEM is particularly well-suited because it accommodates reflective measurement models, handles non-normal data distributions effectively, and is appropriate for studies with moderate sample sizes and predictive theoretical purposes.⁷⁰ The study employed a purposive convenience sampling strategy. Given the limited access to a comprehensive national sampling frame for the UAE’s diverse, multi-sector workforce, which comprises 89% expatriates from over 200 countries, a convenience sample was the most feasible approach to reach respondents across multiple sectors within the project timeline. Convenience sampling is widely accepted in organizational psychology research when the primary objective is theory testing rather than population-level generalization,⁷¹ and is consistent with similar JD-R studies.^61,66 The sample size of N = 160 exceeds the minimum threshold recommended for detecting medium effect sizes (f² ≥ 0.15) with 80% power at α = 0.05, which requires approximately 77 observations for the maximum number of predictors in any single equation.⁷⁰

Sample and data collection

Data were collected via a structured online questionnaire distributed through professional networks and organizational contacts across various sectors in the UAE, including education, healthcare, manufacturing, construction, and transportation. Table 4 below presents the demographic profile of respondents. All participants were required to provide written informed consent prior to participation, with responses anonymized to ensure confidentiality. The questionnaire yielded a final valid sample of N = 160 respondents.

Measurement instrument

The questionnaire operationalized six latent constructs using multi-item reflective scales measured on a five-point Likert scale (1 = Strongly Disagree to 5 = Strongly Agree). Job resources (JR1–JR10) and personal resources (PR1–PR4) were combined into a single second-order “Resources” construct consistent with Bakker et al.,⁵⁸ who treat job and personal resources as functionally equivalent energetic regulators within the JD-R framework, both serving to buffer demands and enable proactive behavior. Table 5 presents the measurement instrument constructs and their respective indicators.

Data analysis

Data analysis followed the two-step PLS-SEM procedure recommended by Hair et al.,⁷⁰: (1) assessment of the measurement model (outer model) to establish reliability and validity, followed by (2) evaluation of the structural model (inner model) to test hypothesized paths. Bootstrapping with 5,000 iterations was employed to generate path coefficient significance levels (t-statistics and p-values) at a two-tailed significance level of α = 0.05. All analyses were conducted using SmartPLS 4⁷² with standardized results reported. The descriptive statistics are shown in Table 6 below.

Measurement model assessment

Prior to evaluating the structural relationships, the psychometric properties of the measurement model were assessed using indicator reliability (outer loadings), internal consistency reliability (Cronbach’s Alpha and Composite Reliability), convergent validity (average variance extracted [AVE]), and discriminant validity (Fornell–Larcker criterion and HTMT ratio).

Construct Reliability and Validity

A summary of construct reliability and validity is shown for all constructs in Table 7.

Indicator Reliability (Outer Loadings)

All indicator outer loadings were assessed against the recommended threshold of ≥0.60.⁷⁰ JR1 was excluded prior to final analysis as its loading (0.55) fell below this threshold. Table 8 presents the outer loadings from bootstrapping with 5,000 subsamples.

Discriminant Validity: Fornell–Larcker Criterion

Table 9 presents discriminant validity results using the Fornell–Larcker criterion.⁷⁴ Diagonal values (√AVE) exceed all off-diagonal correlations, confirming discriminant validity.

Discriminant Validity: Heterotrait-Monotrait (HTMT) Ratio

The HTMT ratio was assessed as an additional criterion for discriminant validity.⁷³ Values below 0.90 indicate acceptable discriminant validity. Table 10 presents the HTMT matrix.

Structural model assessment

As shown in Figure 2, the measurement model’s acceptability was confirmed; the structural model was assessed using PLS-SEM bootstrapping (5,000 subsamples, two-tailed tests, α = 0.05) to estimate path coefficients, effect sizes (f²), and predictive relevance (Q²).

Figure 2. Structural Equation Model results from SmartPLS 4⁷² (PLS-SEM Bootstrapping, N = 5,000).

Path coefficients (β) are displayed on each structural arrow. Significant paths (p < 0.05) are shown in solid lines; non-significant paths are shown in dashed lines. R² values: Job Crafting (R² = 0.544), Work Engagement (R² = 0.336), Exhaustion (R² = 0.097), Job Performance (R² = 0.630).

Path Coefficients

Table 11 presents the Path coefficients results.

R ² Values and Model Explanatory Power

Table 12 presents the R² values of the constructs.

Effect Sizes (f ²)

Table 13 presents the Cohen’s effect size of all the paths.

Predictive Relevance (Q ²) and Model Fit

Table 14 presents the Predictive relevance and model fit indices of the research.

Indirect and Total Effects

Table 15 presents the total and specific indirect effects.

The pivotal role of job crafting (H1, H5, H7)

The most striking finding is the dominant role of job crafting in driving performance. H1 (Resources → JC: β = 0.759, f² = 1.180) is the strongest path in the model, aligning closely with Conservation of Resources (COR) theory,^78–80 which posits that individuals with abundant resources are motivated to proactively invest them to acquire additional resources. This “gain spiral” manifests in the UAE context as resource-rich employees actively reshaping their roles, relationships, and task boundaries.⁸¹ H5 (JC → WE: β = 0.455, f² = 0.146) and H7 (JC → JP: β = 0.727, f² = 0.939) further confirm the centrality of job crafting: not only does it substantially enhance engagement, but it is also the most powerful direct predictor of job performance in this model. This finding extends the work of Rudolph et al.⁸² and Dubbelt et al.⁸³ to a UAE context, demonstrating that proactive job redesign behaviors transfer robustly across cultures. The indirect pathway Resources → JC → JP (β = 0.552, CI [0.422, 0.678]) explains the dominant mechanism through which resources translate into performance, with job crafting acting as a full mediator.

The health-impairment pathway (H3) and the non-significant downstream effect (H9)

H3 (JD → EX: β = 0.281, f² = 0.087, p = 0.005) confirms the classical JD-R health-impairment pathway: elevated job demands deplete employees’ energy, resulting in exhaustion. This is consistent with the broader empirical literature^61,63 and is particularly relevant in the UAE context, where 55% of respondents work 40–49 hours weekly. However, H9 (EX → JP: β = −0.014, f² = 0.000, p = 0.790) was not supported, indicating that exhaustion does not directly translate into reduced job performance in this sample. This finding departs from classical burnout literature but resonates with the cultural dynamics of the UAE workforce. As a predominantly expatriate labor market where job retention concerns are acute, UAE employees may demonstrate remarkable resilience in maintaining performance standards even under conditions of emotional depletion consistent with the professionalism norms documented by Khan & Khurshid³⁷ and Almarzooqi et al.¹⁶

Non-significant paths: H2, H4, H6, H8

Four hypotheses were not supported, each revealing substantive contextual insights. H2 (Resources → WE: β = 0.157, p = 0.094) was not directly significant; however, the strong indirect effect Resources → JC → WE (β = 0.345, p < 0.001) demonstrates that in the UAE context, resources do not independently fuel engagement without first triggering crafting behaviors. H4 (JD → JC: β = 0.121, p = 0.072) was marginally non-significant, potentially reflecting UAE-specific cultural appraisal of demands as hindrances rather than challenges. H6 (JC → EX: β = −0.118, p = 0.199) was also not supported, suggesting that while crafting enhances engagement, it does not significantly reduce exhaustion. This may reflect the dual-edged nature of crafting: increasing challenging demands may also increase workload, partially offsetting restoration benefits.⁶⁸ Finally, H8 (WE → JP: β = 0.106, p = 0.096) was marginally non-significant, suggesting engagement’s impact on performance may be contingent on UAE-specific contextual and measurement factors.

Comparison with prior literature

These findings broadly support and extend the JD-R literature^58,63 while revealing important UAE-specific contextual boundary conditions. The dominant role of job crafting aligns with Tims et al.⁸¹ and Rudolph et al.⁸² but reveals that the crafting-engagement-performance chain is the primary performance driver in this context. The non-significant exhaustion–performance relationship contrasts with findings from Western studies^61,66 but is consistent with the dynamics of cultural professionalism and occupational precarity documented in UAE-specific literature.^16,37 The model explains 63.0% of the variance in job performance, a substantial R² value that underscores the explanatory power of the JD-R framework in this setting.

Summary of findings

This study successfully applied PLS-SEM to test the JD-R model in a UAE organizational context (N = 160). The key findings are:

Practical recommendations

The model paths reported are strongly aligned with established theories within the Conservation of Resources (COR) and Job Demands–Resources (JD-R) frameworks. Practical recommendations include:

Conclusions

This study provides four key conclusions for the UAE organizational context:

Practical implications