Optimizing Multidisciplinary Workforce Sustainability: The Interplay of Job Strain, Workplace Hazards, and Organizational Support on Work Engagement Among Registered Nurses and Practical Nurses and Nursing Assistants

Introduction

Work-related stress is defined by the World Health Organization as a response to work demands and pressures that exceed an individual’s knowledge, abilities, and capacity to cope.¹ It stems from interrelated factors, including the nature of work, organizational management, and the workplace environment.² While optimal pressure can enhance motivation and productivity, excessive stress surpasses individual thresholds, adversely affecting employee health and job effectiveness.^2–4

Healthcare professionals are highly vulnerable to occupational stress due to daily clinical responsibilities and persistent exposure to critical patient situations.⁵ Key predictors include heavy workloads, inadequate resources, workplace conflicts, and low job satisfaction.⁶ Within this sector, the nursing profession encompasses diverse roles such as registered nurses (RNs) and practical nurses and nursing assistants (PNNAs), distinguished by their scopes of practice and training durations.⁷ Despite working within the same multidisciplinary setting, the occupational descriptions of RNs and PNNAs might exhibit varying levels of work stress. Registered nurses have more specific scopes of practice and longer training periods, which might face higher psychological distress.^8,9 In contrast, nursing assistants, who primarily provide supportive care, are documented to experience greater physical labor pressures, including high frequencies of manual patient handling and repetitive physical tasks.¹⁰ Such occupational stress compromises service quality, decreases patient satisfaction, and increases turnover.¹¹ The workplace constitutes a statistically significant source of stress for employees owing to excessive workloads, psychological harassment, work procedures, encounters with patients’ families, professional and administrative obligations, resource limitations, and insufficient management support.¹² Work-related stress can also lead to burnout, diminished job satisfaction, and reduced work engagement among employees within the organization.^13,14 Work engagement is characterized as a positive and fulfilling psychological state related to work, encompassing three dimensions: vigor, which reflects energy; dedication, indicating involvement; and absorption, representing a state of happiness in work activities.¹⁵ Existing literature indicates that diminished work engagement correlated with a decline in the safety and quality of clinical operations, encompassing lower levels of care quality, decreased patient satisfaction, heightened adverse events, and a contagion effect on patients. The operational and financial viability of healthcare companies was related to heightened turnover intentions, absenteeism, lateness, and negative cultural contagion.^16,17

Tertiary hospitals represent high-risk environments for stress due to their complex, multi-departmental structures¹⁸ making nursing a profession highly susceptible to compassion fatigue.¹⁹ While work strain and low engagement are global challenges, these issues are notably prevalent within the healthcare systems of Southeast Asia. Studies conducted in the Philippines and Vietnam investigated job-related aspects and work engagement among nurses, revealing concerns associated with administrative components and their consequent effects on patient care outcomes.^20,21 In Thailand, empirical evidence indicates that approximately 15–20% of hospital personnel experience significant job stress, which is often associated with critical work factors such as nursing shortages, extended shifts, insufficient income, and poor social support.^22,23 Investigating a multidisciplinary nursing workforce in Thailand addresses a critical regional gap where empirical evidence is limited. Thus, this study examines the prevalence of occupational stress, degrees of work engagement, and the relationship between these variables among healthcare professionals in a medical university hospital.

The study’s theoretical framework is grounded in Karasek’s model of occupational stress, which balances psychological demands against decision latitude,²⁴ to assess diverse occupational risks. Additionally, the Job Demands-Resources theory conceptualizes work engagement as a distinct state independent of burnout.^25,26 Work engagement should be regarded solely as being separate from burnout. This concept posits that, theoretically, an employee who is not experiencing burnout may exhibit either high or low levels of engagement, whereas an engaged employee may demonstrate either high or low levels of burnout. We evaluate occupational stress and work engagement through questionnaires rooted in a theoretical framework to investigate and clarify the association between these variables among study participants.

Materials and Methods

Study Design and Participants

This cross-sectional study was conducted via an on-site survey from 15 April 2023 to 31 May 2023. This study utilized a total population sampling approach, where all eligible RNs and PNNAs across all clinical and administrative departments of the participating university hospital were invited to participate. Of the total distributed questionnaires, a response rate of approximately 30% was achieved. A survey was distributed to 2329 RN and PNNAs who have been employed as the hospital permanent workers for at least 6 months. The sample size determination reveals that a minimum of 532 samples must be assembled, according to calculations by the n4Studies application regarding finite sample mean estimation at Confidence interval was set to 95% (alpha 0.05) without cluster sampling with a pre-specified Utrecht Work Engagement Scale.

(UWES-17) score consisting of standard deviation of 1.30 and absolute error (d) of 0.1. Of 883 responses, 694 (78.6%) participants completed all the questions. The two target occupational groups within the multidisciplinary workforce are defined based on national healthcare regulations in Thailand. RN possess a minimum of a four-year bachelor degree in nursing, are licensed by the Thailand Nursing and Midwifery Council, and are legally responsible for advanced clinical judgment, patient care management, and administrative supervision. The secondary comparative group is collectively designated as PNNAs. Practical nurses complete a formalized one-year certificate program in practical nursing under accredited institutions, operating within a structured supportive clinical scope under the supervision of registered nurses. Nursing assistants typically undergo short-term vocational training lasting from three to six months, focusing primarily on basic patient hygiene, ambulation assistance, and routine physical labor tasks. These distinct educational requirements and operational boundaries underpin the rationalization of evaluating variations in job strain, decision latitude, and workplace hazards between the two cohorts.⁷

Questionnaire

The participants were asked about their Socio-demographic, working conditions, stress from work, and work engagement. The questionnaire consisted of four sections:

Part 1: Socio-demographic of the participants such as age, gender, education, marital status, number of children, financial status, physical and mental health conditions, and average nightly sleep duration.

Part 2: Working conditions such as job position, type of employment, work experience, patient contact.

Part 3: Thai Job Content Questionnaire (JCQ) 54 items, developed by Pitchaya Parkthongsuk et al.^27,28 To ensure linguistic and cultural appropriateness, the research instruments underwent a rigorous adaptation process. The original English scales were translated into Thai using a standard forward-and-backward translation procedure by independent bilingual translators. Subsequently, the Thai version was evaluated for content validity and cultural relevance by a panel of nursing and healthcare management experts. Items were refined based on expert consensus to ensure that the terminology aligned with the daily practice and organizational structure of Thai university hospitals. The questionnaire is divided into 6 areas:

1. Job control refers to the perception of one’s ability to control work and the freedom to make decisions regarding various work-related matters (Questions 1–11).
2. Regarding psychological job demands, it refers to the mental perception related to the responsibilities of performing the job (Questions 12–23).
3. Regarding physical job demands, it refers to the physical perception related to the responsibilities of the job (Questions 24–29).
4. Regarding job security, it refers to the perception of stability in the job being performed (Questions 30–34).
5. Social support refers to the perception of support in job performance from supervisors (supervisor support) and co-workers (co-worker support) (Questions 35–42).
6. Hazard at work refers to the perception of risks or dangers that may arise from performing tasks (Questions 43–54).

The questionnaire predominantly comprises a Likert scale. Questions 1–42 utilized four Likert scales as follows. Strongly agree (4 points), agree (3 points), disagree (2 points), and strongly disagree (1 point). For questions 43–54, which are evaluations of hazards or occupational risks, the responses are classified into three tiers. The three levels are “no problem,” “some/a few problems,” and “many problems,” corresponding to scores of 1, 2, and 3, respectively.

Assessing occupational stress and the prevalence of work-related stress in accordance with the Job Demand-Control Model. The model created by Karasek will take into account the scores derived from job control or decision-making authority (questions 1–11) and psychological job demand (questions 12–23). By dividing the scores obtained from these two components into 2 groups, high and low, using the cut point at the 50th percentile. We utilized the established cut off threshold reported the previous study in similar setting. The JCQ were categorized as high in the following cut off threshold. Job control (≥17), psychological demand (≥34), physical demand (≥15), job security (≥16), social support at work (≥36) and hazards at work (≥17).²⁹ Can assess work-related stress in four types as follows.

1. Low-strain job or relaxed job: high job control low psychological job demand
2. Passive job: low job control, low psychological job demand
3. Active job: high job control, high psychological job demand
4. High-strain job: low job control, high psychological job demand

The content of this questionnaire section has already passed the validity test. Internal consistency reliability: Cronbach’s alpha coefficient was adequate for decision latitude scale (0.82), psychological demand (0.76), social support (0.81), physical demand (0.71), and work hazards (0.86) but moderately adequate for job security (0.55). Permission was obtained from the original developers to include it in the study’s questionnaire.^27,28

Part 4: Work engagement: The Utrecht Work Engagement Scale, Thai language version, (UWES-17-TH). The UWES, consisting of 17 items, uses a Likert-type response format with scores ranging from 0 (never) to 6 (always, every day). Three subscales are included: vigor (6 items), dedication (5 items), and absorption (6 items).³⁰ The Utrecht Work Engagement Scale, Thai language version, has already passed the validity test, with Cronbach’s alpha coefficient for overall work engagement as 0.93, absorption as 0.79, dedication as 0.88, and vigor as 0.84.³¹ The interpretation of the tool used will use the median as the criterion. If the score is lower than the median, it will be categorized as low. If the score is higher than or equal to the median, it will be categorized as high.

Statistical Analysis

All statistical analyses were conducted using statistical packages in the Stata software version 18.0 (StataCorp, TX, USA). Descriptive statistics were provided by frequency and percentage for categorical data, a mean with standard deviation (SD) for the parametric data and a median with interquartile range (IQR) for the non-parametric data. One-sample Kolmogorov–Smirnov test of normality was used to assess the outcome distribution for normality including UWES-17 and three subdomains. The non-parametric rank-sum test (Mann–Whitney U-test) was utilized to evaluate the statistical significance of between-group differences in total work engagement sum scores and its respective subdomains (vigor, dedication, and absorption) between RNs and PNNAs. Exploratory analysis using a multivariable linear regression with robust standard error correction was performed to determine factors associated with UWES-17 sum of scores and its subdomains: vigor, dedication, and absorption. The paper questionnaire was distributed through the head of each department office and manually entered by trained personnel into spreadsheet software by two person and cross-validate for any discrepancies. All discrepancy was traced back to hardcopy and resolved for any typographic error in data entry. Extreme values were prevented by specifying ranges for ages (20–60 years old), sleep (2–12 hours) in both data gathering and entry process; value outsides the range were treated as missing values. No missing data imputation was made due to missing completely at random assumption (MCAR), identified using sensitivity analysis of samples: 1) complete response (n = 506) and 2) partially completed response across three subdomains of UWES-17 (n = 516 to n = 518). The results of this study were reported in accordance with the strengthening of the reporting of observational studies in Epidemiology (STROBE) checklist. All statistical analyses were two-sided, and a p-value ≤ 0.05 was considered statistically significant.

Ethical Considerations

This study was conducted in accordance with the Declaration of Helsinki guidelines, and the protocol was approved by the Research Ethics Committee, Faculty of Medicine, Chiang Mai University, Thailand (No. 120/2023). All participants read the Informed Consent form carefully and provided their consent before starting to complete the questionnaire. The study objectives and procedures were clearly explained, and participants were informed of their right to withdraw at any time.

Results

The majority of participants were women (636, 91.6%), held a bachelor’s degree (460, 66.3%), were single (416, 59.9%), and had no children (447, 64.4%). Most reported no underlying physical (469, 67.6%) or psychiatric (676, 97.4%) conditions and described their financial status as more than adequate (360, 51.9%). The average sleep duration was 6 hours per night, and mean work experience was 13 years. Statistically significant differences between subgroups were observed in marital status (p = 0.011), parenthood (p = 0.013), and sleep duration (p = 0.002). Highly statistically significant differences were found in education level, financial status, and employment type (p < 0.001) (Table 1).

Table 1 Sociodemographic Characteristics and Working Conditions of Eligible Participants

Participants reported low job control (277, 39.9%), high psychological demand (423, 61%), high physical demand (434, 62.5%), low support (146, 21%), and high job hazard (358, 51.6%) (data not shown). Comparison of Job Content Questionnaire results showed statistically significant differences in psychological demand, physical demand, and job hazard between RN and PNNAs groups (Figure 1).

Figure 1 The descriptive bar plots demonstrate subdomains of the 54-item Job Content Questionnaire (JCQ) across occupation: 1) registered nurses (gray) and 2) practical nurses and nursing assistants (Orange); descriptive statistics presented in frequency and percentage, n (%); p-value was derived from the Chi square test of two independent proportions.

Using the job control-demand matrix, most participants (253, 41.3%) were classified as having an Active job, followed by High-strain (157, 25.7%), Relaxed (127, 20.8%), and Passive (75, 12.3%) jobs (data not shown). The detailed distribution and descriptive parameters of occupational categories for both RN and PNNAs are presented comprehensively in Figure 2.

Figure 2 The descriptive mosaic plots demonstrate four phenotypes of job control-demand matrix: 1) active (blue), 2) high-strain (red), 3) relaxed (green), and 4) passive (gray); descriptive statistics presented in frequency and percentage, n (%); (A) registered nurses (n = 506); (B) practical nurses and nursing assistants (n = 188).

For work engagement, One-sample Kolmogorov–Smirnov test of normality was used on the data of UWES-17 and three subdomains resulting in presence of non-normality (p < 0.01). PNNAs exhibited a significantly higher overall work engagement sum score compared to RNs (Median: 76 (Q1=59, Q3=87) vs 67 (Q1=51, Q3=82), p < 0.01). Consistent with the global score, statistically significant differences were observed across all three subdomains. Specifically, PNNAs reported higher scores than RNs in vigor (Median: 28 (Q1=24, Q3=31) vs 27 (Q1=21, Q3=30), p = 0.02), dedication (Median: 21.5 (Q1=14.5, Q3=26) vs 18 (Q1=12, Q3=24), p < 0.01), and absorption (Median: 25 (Q1=18, Q3=30) vs 23 (Q1=16, Q3=28), p < 0.01). Between-group comparisons showed statistically significant differences in overall engagement, dedication, and absorption (Figure 3).

Figure 3 The descriptive bar plots demonstrate subdomains of the 17-item Utrecht Work Engagement Scale (UWES-17) across occupation: 1) registered nurses (gray) and 2) practical nurses and nursing assistants (Orange); descriptive statistics presented in frequency and percentage, n (%); p-value was derived from the Chi square test of two independent proportions.

Multivariable linear regression identified factors associated with work engagement and its subdomains (Tables 2 and 3). High-strain jobs were linked to lower engagement across all domains in both groups, while high support was associated with higher engagement and vigor. Among PNs, older age, high job hazard, and high physical demand were associated with lower engagement or dedication. In the PNNAs group, high job hazard reduced absorption scores, whereas high support increased dedication scores. The completed response analysis is presented in Tables S1 and S2.

Table 2 Multivariable Linear Regression Demonstrating the Association of Determinants with Work Engagement Score and Its Subdomains in Registered Nurses

Table 3 Multivariable Linear Regression Demonstrating the Association of Determinants with Work Engagement Score and Its Subdomains in Practical Nurses and Nursing Assistants

Discussion

The study’s first objective revealed that the RN study population predominantly occupied active and high-strain jobs, collectively comprising over half of the RN cohort. This indicates a higher proportion of high-strain jobs compared to previous findings in similar and different Thai hospital contexts, including urban tertiary and private institutions.^29,32,33 Conversely, the PNNAs group was characterized mostly by relaxed and passive jobs. Despite sharing direct patient care responsibilities, considerable disparities in job characteristics are obvious between these populations. These RN job characteristics align with prior research indicating that RNs rank highest among professions for job stress.³⁴ Additionally, different sleep durations between groups are associated with work-related stress, as inadequate sleep relates to physiological stress in the workplace.³⁵

Regarding work engagement, over half of all participants exhibited a high overall level, though the proportion was lower within the RN group than the PNNAs group. Statistically significant differences were found in the proportions of high total, dedication, and absorption scores between the two cohorts. This high engagement level aligns with previous evidence among health professionals.¹³ The distinct dedication and absorption scores in the RN group compared to the PNNAs group relate to differing inherent decision-making processes,³⁶ since decision latitude is associated with dedication, which in turn links to absorption.³⁷

The secondary objective investigated the relationship between work stress and engagement. In both cohorts, high-strain jobs significantly associated with lower scores in overall work engagement and all subdomains. Elevated demands in high-strain jobs serve as a significant mediating factor for adverse health outcomes, with burnout subsequently demonstrating a negative association with engagement.^38,39 Conversely, high support significantly associated with increased overall engagement and vigor, consistent with studies showing job demand and control positively correlate with engagement.^40,41 Within the various socio-demographic and professional traits linked to engagement,⁴² only age and sleep hours showed significant associations among RNs. This aligns with research proving age positively correlates with nurses’ engagement.^21,43 Conversely, sleep duration lacked sufficient evidence to establish a direct link with engagement; only one study indicated that Daylight Saving Time linked to reduced engagement through decreased sleep quality rather than length.⁴⁴ Nonetheless, sleep duration reflects overall well-being and quality of life,⁴⁵ both positively connected to engagement.⁴² The greater sleep amount observed here may suggest improved sleep quality, contributing to a positive relationship with engagement among RNs.

Our findings indicate high physical demands negatively associated with RNs’ dedication, showing that elevated physical burdens coexist with diminished professional dedication. Additionally, high support significantly correlated with nearly all engagement dimensions, except for dedication and absorption among PNNAs. High workplace hazards linked to almost all engagement dimensions in RNs, except for a marginal correlation with absorption; conversely, in PNNAs, hazards significantly associated with absorption. These variations are linked to distinct hazards, psychological demands, physical demands, and workloads between groups, as heavy workloads and hazards demonstrate strong associations with diminished engagement.⁴⁶ Workplace hazards affect work engagement by depleting employees’ cognitive and emotional resources, prompting a transition from a focus on development to self-preservation. Psychological theories indicate that adverse situations serve as stressors, increasing burnout and diminishing job satisfaction.^47,48 Furthermore, physical demands involving non-ergonomic activities, such as prolonged high-risk procedures, heavy lifting, or awkward postures^49,50 explain why participants facing high physical burdens exhibit poor overall dedication. To improve engagement, strategies must focus on reducing stress, optimizing physical demands, minimizing hazards through occupational monitoring, and strengthening team support. Additionally, implementing Decision Support Systems can mitigate psychological demands by addressing communication and decision-making obstacles.^51–53

These factors relate to varying individual perceptions of job demands⁵⁴ alongside organizational factors where disparate management systems connect with employee perceptions.⁵⁵ Given the complexity of examining work involvement via institutional frameworks, future applications of advanced theories, such as the Person-Situation Approach, Multilevel JD-R Theory, or healthcare-specific frameworks like Convergent Care Theory, may yield deeper insights into the association between occupational stress and work engagement.^56,57

This study’s strength lies in its pioneering exploration of psychosocial outcomes among healthcare personnel working in the complex context of a medical university hospital. This research offers insights into work-related stress levels and work engagement, defined in detail in each module, and focuses on comparing specific job groups, in contrast to other studies that focus on a broad workforce. Furthermore, our findings can be utilized to formulate policies and cultivate an organizational culture that enhances employee job engagement. Several limitations must be acknowledged when interpreting this study. First, the cross-sectional design precludes establishing causality and introduces potential reverse causality. For instance, staff with lower baseline engagement might experience decreased situational awareness, increasing their susceptibility to hazards and administrative strain over time. Future longitudinal research is required to clarify these temporal directions. Second, profession heterogeneity and non-random recruitment mean the sample may not fully represent the institutional population. Notably, the low response rate (29.8%) introduces potential non-response and self-selection bias. Overworked or severely burned-out staff likely lacked time to participate, potentially biasing work engagement estimates upward. Therefore, these descriptive findings warrant cautious interpretation. Third, relying entirely on self-report questionnaires (JCQ and UWES) subjects the findings to information, social desirability, and common method biases, which may distort the true relationships among variables. Fourth, this study lacked individual-level data on shiftwork configurations, failing to fully isolate scheduling-related stress. In typical Thai tertiary hospitals, RNs and PNNAs predominantly rotate through continuous 8-hour shifts (morning, afternoon, night). Because shift rotation strongly correlates with fatigue, this omission is a limitation. Residual confounding from unmeasured variables (eg., baseline burnout) may also persist, necessitating detailed scheduling parameters in future research. Finally, the single-hospital, single-country design limits generalizability to other healthcare tiers, private sectors, or different cultural contexts. Future studies should broaden their scope to explore wider predictors, utilize advanced analytical approaches, and investigate concurrent mental health conditions like anxiety and depression.

Conclusion

This study highlights marked differences in stress and engagement between RNs and PNNAs; RNs reported higher demands, while PNNAs showed greater engagement. Across both groups, high-strain jobs reduced engagement, whereas high support increased it. Importantly, regression analysis identified organizational support as the primary intervention target with the most significant impact, followed by job strain, sleep duration, and workplace hazards. Consequently, interventions must be tailored for practical operability. For RNs, strategies should optimize workflows and establish peer-support programs to reduce decision-making pressure. For PNNAs, priority must be given to ergonomic training and safety protocols to mitigate physical labor. Regarding generalizability, these findings offer substantial reference value for comparable secondary and tertiary hospitals across Thailand. Despite cross-sectional limitations, these insights effectively inform policies for a healthier, more engaged healthcare workforce.