*Article* **Predictors of Burnout in Healthcare Workers during the COVID-19 Pandemic**

**Adriana Cotel 1,†, Florinda Golu 1,†, Anca Pantea Stoian <sup>2</sup> , Mihai Dimitriu 3,4 , Bogdan Socea 5,6,\* , Catalin Cirstoveanu 7,\* , Ana Maria Davitoiu <sup>7</sup> , Florentina Jacota Alexe <sup>4</sup> and Bogdan Oprea <sup>1</sup>**


**Abstract:** The purpose of this study was to identify the predictors of burnout in healthcare workers during the COVID-19 pandemic. Data were collected from March to June in 2020, during the COVID-19 pandemic, from employees of two Romanian hospitals. Five hundred and twenty-three healthcare workers completed a series of questionnaires that measured burnout, job demands, job resources, and personal resources. Among the respondents, 14.5% had a clinical level of exhaustion (the central component of burnout). Three job demands (work–family conflict, lack of preparedness/scope of practice, emotional demands), three job resources (training, professional development, and continuing education; supervision, recognition, and feedback; autonomy and control), and one personal resource (self-efficacy) were significant predictors of burnout, explaining together 37% of the variance in healthcare workers' burnout. Based on our results, psychological interventions during the COVID-19 pandemic for healthcare employees should focus primarily on these demands and resources.

**Keywords:** burnout; COVID-19; health personnel; pandemics

#### **1. Introduction**

The outbreak of Coronavirus Disease 2019 (COVID-19) is considered a global health threat [1], becoming the third major coronavirus outbreak in recent times following severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) [2]. The challenges associated with the COVID-19 pandemic (e.g., heavy workload, work pressure, high risk of infection, inadequate resources) may affect the mental health of medical staff, such as frontline workers, mainly in terms of their burnout level [3]. Burnout represents a job-related strain as a result of repeated exposure to stressors at work, which is characterized by exhaustion (i.e., the depletion of one's emotional and physical resources), cynicism (i.e., the negative detachment form work), and reduced efficacy (i.e., the perception of one's lack of productivity and achievement) [4]. During the outbreak of COVID-19, the prevalence of burnout among healthcare workers ranged between 13% and 51%, depending on the country, the specific job in the hospital, and the period in which the data were collected [5–8]. However, there are insufficient data regarding the predictors of burnout

**Citation:** Cotel, A.; Golu, F.; Pantea Stoian, A.; Dimitriu, M.; Socea, B.; Cirstoveanu, C.; Davitoiu, A.M.; Jacota Alexe, F.; Oprea, B. Predictors of Burnout in Healthcare Workers during the COVID-19 Pandemic. *Healthcare* **2021**, *9*, 304. https:// doi.org/10.3390/healthcare9030304

Academic Editor: Alberto Modenese

Received: 15 February 2021 Accepted: 8 March 2021 Published: 9 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

during the COVID-19 pandemic. Identifying these predictors is important in order to develop the best individual and organizational interventions that could provide support for medical staff during the COVID-19 pandemic and during possible future pandemics. Since the burnout of healthcare staff is associated with decreased quality of care and decreased safety of patients [9], the efficient management of medical personnel's burnout has practical implications for both the employees in the medical sector and for the patients, with major consequences for how the health system responds to current or future outbreaks.

Previous studies during SARS and MERS outbreaks indicate a series of job characteristics that are related to burnout in healthcare employees. After the 2003 outbreak of SARS, healthcare workers from hospitals that treated SARS patients reported higher levels of burnout than hospital employees who had not treated such patients [10]. Their perceived adequacy of training, protection, and support was negatively associated with burnout [10]. Support from supervisors, colleagues, and the organization was a negative predictor of psychiatric symptoms and of psychological distress during the SARS outbreak [11,12]. The emergency department nurses' burnout was related to the lack of resources for treatment during the MERS outbreak [13].

These findings are in line with the Job Demands–Resources Theory (JD-R) [14], which suggests that job demands lead to a higher level of burnout and job resources decrease burnout. In addition, the theory suggests that personal resources lead to lower levels of burnout. This last assumption of the theory is supported by data from the medical sector. Under normal working conditions, personal resources such as optimism and self-efficacy are related to decreased levels of burnout in nurses [15,16].

Based on the JD-R theory and on previous research during SARS and MERS outbreaks, job demands are expected to be positively associated with burnout, and both job resources and personal resources are expected to be negatively associated with burnout in healthcare workers during the COVID-19 pandemic.

#### **2. Materials and Methods**

Employees of two hospitals from Romania were asked to complete a questionnaire including all the studied variables. The questionnaire was distributed in a paper-and-pencil form or in an online form. The questionnaires were distributed through the managers or the decision-makers of the hospitals. They were contacted and informed about the purpose of the study and were asked for permission to collect data. They received the hard copy questionnaires or the link to the online form and were asked to distribute them to the employees of the two hospitals. The hard copy questionnaires were collected by one of the researchers after several visits to the hospitals, following safety and protection measures. The managers or the decision-makers did not have access to the participants' answers. The completion of the questionnaire took approximately 30 min. Data were collected from March to June in 2020, during the COVID-19 pandemic. All procedures performed in the study were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. To protect data confidentiality, participants completed questionnaires anonymously, and data were analyzed globally. The questionnaires in hard copy were stored in a safe place, and only those who performed the statistical analyses had access to the online database. Informed consent was obtained from all individual participants included in the study, according to the legal rules of informed consent [17]. Out of a total of 544 responses, 21 were invalidated due to missing values. Table 1 illustrates the characteristics of the two groups and the results of their comparison. There were no significant differences between study participants and those excluded due to missing data in relation to age, gender, tenure, and occupation.


**Table 1.** Comparison between included and excluded participants.

Burnout was measured using the Maslach Burnout Inventory—General Survey [18,19]. The 16 items of the scale measure three components of burnout: exhaustion (5 items, e.g., "I feel burned out from my work."), cynicism (5 items, e.g., "I have become less enthusiastic about my work."), and professional efficacy (6 items, e.g., "I feel confident that I am effective at getting things done."). The items are scored on a 7-point Likert scale, from 0 (never) to 6 (every day).

Job demands were measured with the Job Demands in Nursing Scale [20]. Lack of comfort with working conditions was measured with 4 items (e.g., "I am satisfied with my day-to-day routine".), lack of preparedness/scope of practice was measured with 4 items (e.g., "I do not feel adequately prepared for my area of practice."), and lack of equipment and supplies was measured with 4 items (e.g., "The equipment needed for patient care is poorly maintained".). The instrument uses a scale from 1 (strongly disagree) to 5 (strongly agree) for all these demands. In addition, emotional demands were measured with 4 items developed specifically for health care professions [21]. Employees reported on a scale from 1 (never) to 5 (always) how often they were confronted with death, human suffering, aggressive patients, and troublesome patients. Quantitative demands were measured with 5 items (e.g., "Do you have to work very fast?") [22] on a scale from 1 (hardly ever) to 5 (always). Finally, work–family conflict was measured with Work–Family Conflict Scale [23], composed of 5 items (e.g., "The demands of my work interfere with my home and family life".) on a 7-point (strongly disagree–strongly agree) response scale.

Job resources were measured with the Job Resources in Nursing Scale [20]. Supervision, recognition, and feedback was measured with 4 items (e.g., "I feel validated by my supervisor for a job well done".), training, professional development, and continuing education was measured with 4 items (e.g., "I am able to access an adequate number of in-services or continuing education activities".), staffing and time was measured with 4 items (e.g., "There are enough staff members in my work setting to get the job done".), technology was measured with 4 items (e.g., "I am able to provide better care because of the information systems and technology available to me".), autonomy and control was measured with 4 items (e.g., "My job description is flexible; I am able to modify my daily duties or the type of work that I do".). The instrument uses a scale from 1 (strongly disagree) to 5 (strongly agree) for all these resources. In addition, social support was measured with the Job Demands–Resources Questionnaire [24], using three items (e.g., "If necessary, can you ask your colleagues for help?") on a scale from 1 (never) to 5 (very often).

Personal resources were measured with the Job Demands–Resources Questionnaire [24]. Self-efficacy (e.g., "I can handle whatever comes my way".) and optimism (e.g., "I usually expect the best in uncertain times".) were measured with four items each on a scale from 1 (absolutely wrong) to 4 (absolutely right).

#### **3. Results**

Burnout scores ranged from 0 to 5, *M* = 1.35, *SD* = 0.93; exhaustion scores ranged from 0 to 6, *M* = 2.05, *SD* = 1.31; cynicism scores ranged from 0 to 5.60, *M* = 1.27, *SD* = 1.10; professional inefficacy scores ranged from 0 to 5, *M* = 0.94, *SD* = 0.92. Statistical analyses indicated an adequate reliability for the burnout measurement: Cronbach's *α* = 0.89 for the overall score, Cronbach's *α* = 0.84 for exhaustion, Cronbach's *α* = 0.74 for cynicism, and Cronbach's *α* = 0.80 for professional inefficacy. We tested the factor structure of the burnout measure by conducting a series of confirmatory factor analyses using M plus 7.0 [25] in order to investigate the validity of the hypothesized measurement model. The first-order and second-order theoretical models of burnout were compared with the model in which all items loaded on a single factor. The fit indices (χ <sup>2</sup> = 295.64, df = 99, χ <sup>2</sup>/df = 2.99, CFI (Comparative Fit Index) = 0.94, TLI (Tucker–Lewis Index) = 0.93, RMSEA (Root Mean Square Error of Approximation) = 0.06 (CI = 0.05, 0.07), SRMR (Standardized Root Mean Square Residual) = 0.05) for the first-order model (in which the three components of burnout were loaded by their specific items) showed a good fit with the data. Similar results (χ <sup>2</sup> = 295.64, df = 99, χ <sup>2</sup>/df = 2.99, CFI = 0.94, TLI = 0.93, RMSEA = 0.06 (CI = 0.05, 0.07), SRMR = 0.05) were found for the second-order model (in which a general burnout factor was loaded by exhaustion, cynicism, and professional inefficacy, which in turn were loaded by their specific items). An alternative model, in which all items loaded on a single factor, showed poor fit with the data (χ <sup>2</sup> = 639.60, df = 102, χ <sup>2</sup>/df = 6.27, CFI = 0.84, TLI = 0.81, RMSEA = 0.10 (CI = 0.09, 0.11), SRMR = 0.07). The burnout measure showed a good fit because CFI (Comparative Fit Index) and TLI (Tucker–Lewis Index) were above 0.90 [26], the RMSEA (Root Mean Square Error of Approximation) was 0.06, and the SRMR (Standardized Root Mean Square Residual) was lower than 0.08 [27]. Therefore, the burnout measure adopted in our study was valid. Since the cutoff points presented in the Dutch version of the Maslach Burnout Inventory (MBI) manual or recommended in other studies were not able to satisfactorily differentiate between clinical and non-clinical burnout, we used a 3.50 cutoff point for exhaustion in order to minimize false negatives [28]. Based on this cutoff point, 76 (14.5%) of the healthcare workers had a clinical level of exhaustion during the COVID-19 pandemic in Romania. There were no differences between men and women in terms of their burnout level (*p* > 0.05). Additionally, there were no differences between professions regarding the level of burnout (*p* > 0.05).

Table 2 shows the means, standard deviations, reliabilities, and the correlations with burnout (the overall score and the three factors) for the variables included in the study (job demands, job resources, and personal resources). As expected, job demands were positively associated with burnout, and both job and personal resources were negatively associated with burnout.

A three-stage hierarchical multiple regression was conducted in order to predict healthcare workers' burnout during the COVID-19 pandemic based on their job demands, job resources, and personal resources. All the job demands were entered at stage one. A significant regression equation was found; *F*(6, 516) = 27.128, *p* < 0.001, with an *R* <sup>2</sup> of 0.23. Job demands accounted for 23% of the variation in burnout. All the job resources were entered at stage two. A significant regression equation was found; *F*(12, 510) = 20.074, *p* < 0.001, with an *R* <sup>2</sup> of 0.31. Adding job resources to the regression model explained an additional 8% of the variation in burnout during the COVID-19 pandemic. The final model also included personal resources; *F*(14, 508) = 22.487, *p* < 0.001, explaining an additional 6% of the variation in burnout. The final model explained 37% of the variance in healthcare workers' burnout. When all the independent variables were included in the regression model, only three job demands (work–family conflict, lack of preparedness/scope of practice, emotional demands), three job resources (training, professional development, and continuing education; supervision, recognition, and feedback; autonomy and control), and one personal resource (self-efficacy) were significant predictors of burnout in healthcare workers during the COVID-19 pandemic. The regression statistics are presented in Table 3. The results are in line with our hypotheses.


**Table 2.** Means, standard deviations, reliabilities, and correlations with burnout (*n* = 523).

Footnotes: *M* = mean, *SD* = standard deviation, *α* = Cronbach's alpha, \*\* *p* < 0.01, \*\*\* *p* < 0.001.



Table footnotes: \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.00.

#### **4. Discussion**

The results of this study are in line with the JD-R theory [14] and with previous research during SARS and MERS outbreaks. Job demands were positively associated with burnout during the COVID-19 pandemic, as in the case of the MERS outbreak for emergency department nurses [13]. A negative relationship between job resources and burnout was found in both the current study and during the SARS outbreak, when training, protection, and support from supervisors, colleagues, and the organization were negative predictors of psychological stress and burnout [10–12]. Regarding the negative relationship between personal resources and burnout, our results are in accordance with findings under normal working conditions in healthcare [15,16]. Workplace stressors for medical staff has been studied in Romania before but not in pandemic conditions [29]. As far as we know, only one study investigated the burnout of Romanian medical personnel during the pandemic, but it focused only on prevalence [30]. The present study contributes to the development of knowledge related to burnout in the medical field during pandemics by highlighting a number of predictors.

Based on our results, psychological interventions during the COVID-19 pandemic for healthcare employees should focus primarily on three job demands (work–family conflict, lack of preparedness/scope of practice, emotional demands), three job resources (training, professional development, and continuing education; supervision, recognition, and feedback; autonomy and control), and one personal resource (self-efficacy). The existing data support the efficiency of some interventions in reducing burnout. Three types of interventions that reduce exhaustion have been identified: those based on relaxation techniques, those that provide new role-related knowledge and work skills, and those that provide cognitive-behavioral therapy [31]. Moreover, job crafting interventions have a positive effect on the well-being and performance of employees in the medical sector [32]. Finally, self-efficacy can be increased with psychological capital interventions [33]. These types of interventions can be used in order to reduce the effect of the identified predictors on burnout.

This study has a number of limitations. Firstly, the job characteristics during the COVID-19 pandemic were measured with self-report instruments. The collected data do not provide an objective evaluation of actual demands such as lack of preparedness or resources such as supervision. Secondly, the sample consists of Romanian employees, raising concerns regarding the generalizability of our findings to other countries. Finally, the study was cross-sectional; therefore, we cannot draw causal conclusions. Future longitudinal studies could identify predictors of medical staff burnout in other countries and using multiple measurement methods.

#### **5. Conclusions**

This paper contributes to the field by extending the JD-R model's assumptions about predictors of burnout in particular work situations, such as the context of an outbreak for healthcare workers. In line with the model, burnout was associated with high demands and with the lack of job and personal resources, supporting the utility of JD-R in understanding negative psychological states at work during pandemics. Our findings suggest that psychological interventions during the COVID-19 pandemic for healthcare employees should focus primarily on three job demands (work–family conflict, lack of preparedness/scope of practice, emotional demands), three job resources (training, professional development, and continuing education; supervision, recognition, and feedback; autonomy and control) and one personal resource (self-efficacy). In these demanding circumstances, practitioners in the field of occupational health psychology can implement cognitive-behavioral interventions, relaxation techniques, job crafting interventions, psychological capital interventions, and trainings aimed at developing work-related knowledge and skills.

**Author Contributions:** Conceptualization, A.C. and F.G.; methodology, F.G. and A.P.S.; software, F.G. and B.O.; validation, B.S., M.D., and F.J.A.; formal analysis, A.M.D.; investigation, A.C.; resources, A.C. and F.G.; data curation, F.G.; writing—original draft preparation, F.G. and B.O.; writing—review and editing, B.S. and A.P.S.; visualization, C.C.; supervision, C.C. and B.O. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki. The study was approved by the Local Ethical Committee of Sf. Pantelimon Emergency Clinical Hospital, Bucharest, under number 6/04.02.2020 and of Maria Sklodowska Curie Clinical Children Hospital, Bucharest, under number 2/10.03.2020.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study, according to the legal rules of informed consent.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to confidentiality reasons.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


#### *Article* **Work-Related Musculoskeletal Complaints in Surgeons**

**Andreea Luciana Rat,ă 1,2 , Sorin Barac 1,2,\*, Loredana Luciana Garleanu <sup>3</sup> and Roxana Ramona Onofrei <sup>4</sup>**


**Abstract:** The aim of the present study was to examine the prevalence of work-related musculoskeletal complaints and potential risk factors among Romanian surgeons. Ninety-five surgeons of different specialties (62.11% males) completed a questionnaire about work-related musculoskeletal complaints (WMSCs). Ninety-one surgeons (95.78%) experienced WMSCs at least in one body part in the last year. Most surgeons reported pain in four body parts (33.68%). The most common WMSCs were reported on the lower back (74.73%), followed by complaints in the neck region (55.79%), shoulder and upper back (46.32%), knee (31.58%), wrist–hand (16.84%), elbow (14.74%), hip (11.58%) and ankle–foot (4.21%). Surgeons rated their pain more severe on upper back, lower back and knees. A higher percentage of male surgeons reported upper back pain (χ 2 (1) = 5.818, *p* = 0.015). Significant age differences were found between the reported pain sites (F8,278 = 2.666, *p* = 0.008); the surgeons reporting wrist–hand pain were younger than those reporting neck, shoulders, elbows, dorsal and lumbar pain. Surgeons with significantly less experience in years reported significantly more WM-SCs in wrist–hand, hip and ankle–foot regions compared with those more experienced (*p* < 0.05). Surgeons are at high risk of developing work-related musculoskeletal complaints, which affects both their professional and personal life. Further studies are needed to identify all risk factors and ergonomic strategies to reduce the prevalence and the negative impact of WMSCs.

**Keywords:** musculoskeletal complaints; pain; surgeons

#### **1. Introduction**

Surgeons, as all other healthcare workers, are at risk of developing work-related musculoskeletal complaints (WMSCs) and disorders. A substantial number of surgeons suffer from work-related musculoskeletal symptoms that are exacerbated as those surgeons continue to operate [1].

Several studies have reported that surgeons are exposed to intense physical strain while performing different surgical procedures. All well-known risk factors for WMSCs are met during surgical procedures—awkward, prolonged static postures, repetitive movements of upper limbs [2]. Kant et al. [3] have analyzed the most common static working position adopted by surgeons and this implies the head bent forward, the spine bent forward and twisted, the shoulder raised and standing on one leg. Ruitenberg et al. [4] found that surgeons stand 130% longer and performed fine repetitive movements 26 times longer than other hospital physicians. The physical demands were perceived by surgeons as uncomfortable and exhausting, the main reason being the prolonged repetitive movements and the working postures. The authors concluded that the physical demands of performing surgery are a threat to surgeons' physical health, work ability and job performance [4]. Another study also described that the discomfort or symptoms reported by surgeons were attributed to performing any mode of surgery—open, laparoscopic or robotic [5].

**Citation:** Rat,˘a, A.L.; Barac, S.; Garleanu, L.L.; Onofrei, R.R. Work-Related Musculoskeletal Complaints in Surgeons. *Healthcare* **2021**, *9*, 1482. https://doi.org/ 10.3390/healthcare9111482

Academic Editors: Fabriziomaria Gobba and Alberto Modenese

Received: 30 September 2021 Accepted: 27 October 2021 Published: 31 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Yang et al. [6] have analyzed the posture, fatigue and pain across different surgical specialties and procedures. They found that the work posture of surgeons performing open operations was more demanding for neck, trunk and right upper limb compared with other procedures (laparoscopic, endovascular), as measured objectively by a IMU system (inertial measurement units). Of note, 50% of the surgeons participating in that study reported moderate or higher levels of fatigue and clinically significantly neck and lower back pain immediately after the operation. Other studies have also studied the ergonomic risks the surgeons are exposed to during surgeries, using objective ergonomic assessment tools [7–9].

WMSCs have a negative impact on surgeons' professional activities, quality or performance of surgical care. One-third of the surgeons having a physical complaint in the arm or knee felt impaired in their work functioning, and one out of seven surgeons had difficulties coping with their work physical demands due to impairments in their physical well-being [4]. Another study revealed that more than half of the injured surgeons reported at least a minimal impact on their operating performance while recovering from injury [10].

The aim of the present study was to examine the prevalence of work-related musculoskeletal complaints and potential risk factors among Romanian surgeons.

#### **2. Materials and Methods**

#### *2.1. Participants*

A total of 95 surgeons of different specialties completed an anonymous questionnaire focused on musculoskeletal complaints. Eligibility criteria were: (1) surgical experience of at least 1 year; (2) no clinically diagnosed inflammatory musculoskeletal disorders; (3) no history of surgeries in the last 6 months.

This cross-sectional study was carried out in accordance with the Declaration of Helsinki and has been approved by the local Ethics (approval No. 173/03.10.2019). Participation in the study was voluntary. Participants who agreed to participate in the study and met the inclusion criteria read and signed an informed consent.

#### *2.2. Assessments*

The questionnaire was structured in two parts. The first part included questions regarding age, sex, height, weight, working experience, weekly working hours, operating hours per week, number of surgeries/week. The second part included questions about WMSCs, questions that were adapted from the Nordic Musculoskeletal Questionnaire, a validated, repeatable sensitive and useful screening tool [11]. A body diagram was used to help the respondents indicate the affected body parts. If a subject answered with "yes" on the question "Have you had pain in your neck/shoulder/elbow/wrist–hand/upper back/lower back/hip/knee/ankle–foot in the last 12 months?", then the participant had to answer the questions regarding the pain severity on a 10-point visual analogue scale (VAS), duration, treatment, and the impact of pain on daily living and on professional life.

#### *2.3. Statistical Analysis*

Data were analyzed with MedCalc Statistical software version 19.2.1 (MedCalc Software Ltd., Ostend, Belgium). Descriptive statistics (means and standard deviation, median and interquartile range (IQR)), number and percentage) were calculated. The relationship between WMSCs and sex, years of experience, daily working hours and number of treated patients was evaluated using Chi-square and Cochrane's Q tests. In order to compare variables based on the affected regions, a one-way ANOVA with Tukey–Kramer post hoc tests and Kruskal–Wallis with Conover post hoc tests were performed. A stepwise logistic regression was performed in order to identify risk factors for pain in each region. Variables with *p*-values < 0.1 in the univariate logistic analysis were entered in the multivariate logistic analysis. The significance level was set at *p* < 0.05.

#### **3. Results**

Ninety-five surgeons (62.11% males; height 173.1 ± 8.3 cm; weight 76.75 ± 17.99 kg) agreed to participate and completed the questionnaires. Their age ranged between 25 and 64 years with a mean age of 37.56 ± 8.74 years. Participants' characteristics are presented in Table 1. The participants had different surgical specialty—general surgery (*n* = 6, 6.32%), vascular surgery (*n* = 10, 10.53%), plastic surgery (*n* = 14, 14.74%), neurosurgery (*n* = 21, 22.11%), orthopedic surgery (*n* = 12, 12.63%), urologic surgery (*n* = 8, 8.42%), cardiac surgery (*n* = 8, 8.42%), thoracic surgery (*n* = 8, 8.42%), obstetric/gynecologic surgery (*n* = 8, 8.42%).

**Table 1.** Participants' characteristics.


BMI—body mass index.

Ninety-one surgeons (95.78%) experienced WMSCs at least in one body part in the last year. Most surgeons reported pain in 4 body parts (*n* = 32, 33.68%), 19 surgeons (20%) in one region, 14 surgeons (14.74%) in 2 regions, 13 surgeons (13.68%) in 5 regions, 11 (11.58%) in 3 regions and only 2 surgeons (2.1%) in 6 regions. The most common WMSCs were reported on the lower back (*n* = 71, 74.73%), followed by complaints in the neck region (*n* = 53, 55.79%), shoulder and upper back (*n* = 44, 46.32%), knee (*n* = 30, 31.58%), wrist–hand (*n* = 16, 16.84%), elbow (*n* = 14, 14.74%), hip (*n* = 11, 11.58%) and ankle–foot (*n* = 4, 4.21%). Pain and subjects' characteristics based on the affected region are presented in Table 2.

Gender differences were found only for upper back pain, with a higher percentage of male surgeons reporting WMSCs in this area (χ 2 (1) = 5.818, *p* = 0.015). Significant age differences were found between the reported pain sites (F8,278 = 2.666, *p* = 0.008); the surgeons reporting wrist–hand pain were younger elbow pain (*p* < 0.05).

Surgical experience was significantly correlated with neck pain (χ 2 (5) = 43.11, *p* < 0.0001), shoulder pain (χ 2 (3) = 16.36, *p* = 0.001), upper back pain (χ 2 (5) = 20.36, *p* = 0.001), lower back pain (χ 2 (5) = 42.83, *p* < 0.0001) and knee pain (χ 2 (3) = 11.33, *p* = 0.01), with those with less than 10 years of experience reporting more WMSCs. Surgeons with significantly less experience years reported significantly more WMSCs in wrist–hand, hip and ankle–foot regions compared with those more experienced (*p* < 0.05).

Surgeons with shoulder pain reported significantly more operating hours/week than those with pain in the cervical, wrist–hand, upper back and ankle–foot regions (*p* < 0.05).

The analysis of the prevalence of WMSCs by surgical specialties revealed that the elbow, upper back and lower back pain was highest in neurosurgeons, while upper and lower back were also prevalent in plastic surgeons (Table 3).

Data related to pain and its impact on surgeons' activities are presented in Table 4. Pain severity differed significantly across regions (F8,287 = 37.77, *p* < 0.0001). Surgeons rated their pain more severe on upper back, lower back and knees compared with all other sites (*p* < 0.05).

A significantly number of surgeons reported pain in the neck and lower back for more than 30 days in the last 12 months (cervical pain χ 2 (2) = 18.42, *p* < 0.001; lower back χ 2 (2) = 29.18, *p* < 0.001). In regards to shoulder, elbow and wrist–hand regions, significantly more surgeons reported pain with a duration of less than 7 days (shoulder χ 2 (2) = 9.86, *p* = 0.007; elbow χ 2 (2) = 17.71, *p* < 0.001); wrist–hand χ 2 (2) = 26.37, *p* < 0.001).

The impact of pain on professional activities, leisure activities or both differed significantly according to the affected region (χ 2 (8) = 30.65, *p* < 0.001; χ 2 (8) = 55.59, *p* < 0.001; χ 2 (8) = 125.73, *p* < 0.001, respectively). The number of surgeons with neck pain whose professional activities were affected by WMSCs was significantly higher compared with those with elbow, wrist–hand and ankle–foot pain (*p* < 0.05). Leisure activities and both professional and leisure activities were affected in a higher proportion by lower back pain compared to the other affected regions (*p* < 0.05). Neck and knee pain affected both professional and leisure activities in a higher percentage than elbow, hip and ankle–foot pain (*p* < 0.05). The majority of surgeons continued working with pain.

Significant differences were observed in the number of surgeons receiving medical treatment for their WMSCs (χ 2 (8) = 73.78, *p* < 0.001), a higher prevalence of surgeons with lower back pain following medical treatment in the last 12 months compared with those with pain located in other body parts (*p* < 0.05). Four surgeons with lower back pain (5.63%) needed sick leave due to their musculoskeletal complaints.

The logistic regression analysis revealed that a higher BMI was a risk factor for upper back pain (OR-1.1, 95% CI: 1–1.21, *p* = 0.04) and elbow pain (OR-1.14, 95% CI: 1.02–1.27, *p* = 0.01). Being female increased the risk for neck pain (OR-2.63, 95% CI: 1.07–6.44, *p* = 0.03). A sedentary lifestyle proved to be a risk factor for shoulder pain (OR-8.7, 95% CI: 2.99-25.29, *p* = 0.0001) and for elbow pain (OR-5.53, 95% CI: 1.5–20.37, *p* = 0.01). Other risk factors related to professional activities were the number of surgeries performed in the last 6 months for neck pain (OR-1.13, 95% CI: 1.01–1.27, *p* = 0.03) and shoulder pain (OR-1.29, 95% CI: 1.12–1.49, *p* = 0.0003); number of weekly working hours for lower back pain (OR-1.13, 95% CI: 1.04–1.21, *p* = 0.001) and for hip pain (OR-1.09, 95% CI: 1.03–1.15, *p* = 0.002); wearing a lead apron during surgeries for upper back pain (OR-3.66, 95% CI: 1.32–10.08, *p* = 0.01).

*Healthcare* **2021**, *9*, 1482


**Table 2.** Subjects' characteristics related to the affected region.

**Table 3.** Prevalence of WMSCs by surgical specialty.

75


*Healthcare* **2021**, *9*, 1482


**Table 4.** Pain characteristics based on regions.

#### **4. Discussion**

The aim of the present study was to examine the prevalence of work-related musculoskeletal complaints and potential risk factors among Romanian surgeons. The main finding of this study was that the surgeons are at high risk for development of work-related musculoskeletal complaints; 95.78% of the respondent surgeons have experienced WMSCs in at least one body part in the last 12 months. Most surgeons reported pain in 4 body parts (33.68%).

Our findings are in agreement with previous results, suggesting that surgeons are at high risk for developing WMSCs [12–18]. Alnefaie et al. [12] reported that 80% of respondent suffered from musculoskeletal manifestations related to surgery, with back and neck being the most affected parts (71.1% and 59.8%, respectively). Dianat et al. [14] found a prevalence of 77.2% of surgeons reporting musculoskeletal symptoms, with 76% of these with pain or discomfort in more than one body region. In the study of Szeto et al. [13], over 80% respondent surgeons reported experiencing at least one area of musculoskeletal symptoms in the past 12 months, neck region (82.9%) being the most affected, followed by lower back (68.1%), shoulder (57.8%) and upper back (52.6%).

The most affected region in our study was lower back (74.73%), followed by complaints in the neck region (55.79%), shoulder and upper back (each 46.32%), knee (31.58%), wrist– hand (16.84%), elbow (14.74%), hip (11.58%) and ankle–foot (4.21%). Similar results were also reported by Auerbach et al. [19], who identified a prevalence of lower back pain of 62% and neck pain of 59%. Radiculopathy was present in 30% cases with lower back pain and 28% cases with neck pain. Upper limb complaints were also prevalent in their study sample, 49% reporting shoulder pain and 24% rotator cuff symptoms. Other studies reported the neck as the most affected region [2,13,15,16,18]. In a study of 141 surgeons of different specialties, Giagio et al. [2] reported the most frequently affected body regions as being the neck (79%), lower back (75%), upper back (59%), shoulders (51%), and wrist and hand (26%). In the study of Kokosis et al. [16], 94% of the responders (plastic surgeons) have experienced musculoskeletal pain, with neck being the most affected area.

In our study, surgeons reported more intense pain at the upper back, lower back and knee, with a higher number of surgeons experiencing pain for more than 30 days in the last 12 months. Lower back pain had a significant impact on both leisure and professional activities, with 39.44% of surgeons receiving medical treatment to ameliorate WMSCs in the lower back. In all cases, the surgeons continued working with pain. Our findings are in accordance with those published by Dianat et al. [14], who also reported the mean severity rating of the symptoms experienced by the surgeon being moderate and high. In their study, almost half of the surgeons reported disruption of their normal activities due to musculoskeletal symptoms. Davis et al. [10] reported that 40% of the surgeons included in their study sustained a musculoskeletal injury in the workplace during their career, the common injured regions being the back, hand and neck. In half of the cases, the pain lasted more than 1 month, 66% of injuries being attributed to chronic causes as strain from the operating posture.

The high prevalence of musculoskeletal complaints among surgeons could be attributed to the physical demands of prolonged static working positions and postures, repetitive movements of the upper limbs during surgeries, with very fine eye–hand coordination, long working hours [2,5,13,17]. Yang et al. reported that the working posture of surgeons performing open surgeries was demanding for the neck, trunk and right upper extremity [6]. Not only the bad posture, but also the equipment use during surgery could be considered a cause of musculoskeletal complaints [18]. Our results showed that the majority of surgeons reporting WMSCs wore a lead apron during surgery.

We found gender differences only for upper back pain, with a higher percentage of male surgeons reporting WMSCs in this area. Previous studies showed a higher prevalence of symptoms in female surgeons in the neck, shoulders, elbows, hand/wrists, upper back, hips, knees and ankles [14]. Female surgeons have been reported to experience more pain and discomfort in the wrists [18] and to be at higher risk for multisite musculoskeletal pain

than male surgeons [20]. In our study, being female proved to increase the risk for neck pain, maybe as a result of an ergonomic disadvantage of being shorter. Sutton et al. [21] considered that the higher prevalence of symptoms reported by female surgeons in the shoulder area is due to the fact that female surgeons need to accommodate to the operating table height by raising their arms. Similar results were also reported by Adams et al. [22], who hypothesized that being shorter and having less upper body strength place female surgeons at risk of developing musculoskeletal symptoms or disorders.

Our results revealed that wrist–hand pain was more frequent in younger surgeons. Moreover, surgeons with less years of experience experienced more WMSCs in the wrist– hand, hip and ankle–foot regions compared with those more experienced. Neck, shoulder, upper and lower back and knee pain were more frequent in those with less than 10 years of experience. Similar results were reported by Alnefaie et al. [12], who found a higher percentage of surgeons with 5–10 years of practice with musculoskeletal manifestations, and also by Kokosis et al. [16], who found that musculoskeletal complaints started early in the training of plastic surgeons. Mavrovounis et al. [23] specified that musculoskeletal symptoms appeared early in the residency in their study sample of neurosurgeons. Hemal et al. [24] found that finger numbness is more common in junior laparoscopic surgeons than in senior surgeons. This aspect could be explained by the limited experience, which could lead to inefficient practice. Early training regarding surgery ergonomics could be of much help in preventing musculoskeletal complaints in surgeons. Giagio et al. [2] found that career longevity of more than 20 years is a protective factor for WMSCs. On the other hand, it might be expected that older surgeons, with many years of experience, would be at risk for developing musculoskeletal complaints, as a result of cumulative exposure to physical demands and stress during surgeries.

The analysis of the prevalence of WMSCs by surgical specialties revealed that the elbow, upper back and lower back pain was highest in neurosurgeons, while upper and lower back were more frequent also in plastic surgeons.

A higher BMI and a sedentary lifestyle proved to increase the risk of upper back, shoulder and elbow pain. Dianat et al. [14] reported that the prevalence of neck, shoulder and lower back symptoms decreased with more time spent on sport/physical activities.

Our findings are in agreement with other studies which showed that the professional high physical demand (number of surgeries, number of weekly working hours) increases the risk of work-related musculoskeletal complaints [4,13,14,20].

Due to the high frequency of work-related musculoskeletal complaints and the impact on surgeons' professional and personal life, we recommend preventive programs that will raise awareness of the importance of ergonomics, working postures, surgery schedules and active breaks.

The present study has some limitations. The relatively small sample size, not including all surgical specialties, and the retrospective recall of symptoms are some of the limitations. No objective assessment tool was used to identify the musculoskeletal complaints. We have not considered the working positions, and this is an aspect that should be considered in future studies in order to prevent work-related musculoskeletal complaints due to bad posture and positions during surgeries. Another limitation is that we did not study the loupes usage, which can lead to supplementary load of the neck and, consequently, higher pain in that region.

#### **5. Conclusions**

Romanian surgeons are at high risk for work-related musculoskeletal complaints that affect both professional and personal life. Further studies are needed to identify all risk factors and ergonomic and educational strategies to reduce the prevalence and the negative impact of WMSCs.

**Author Contributions:** Conceptualization, A.L.R. and R.R.O.; methodology, A.L.R. and R.R.O.; validation, A.L.R. and R.R.O.; formal analysis, A.L.R. and R.R.O.; investigation, A.L.R.; data curation, R.R.O.; writing—original draft preparation, A.L.R., S.B., L.L.G. and R.R.O.; writing—review and

editing, A.L.R., S.B. and R.R.O.; visualization, A.L.R. and R.R.O. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of "Pius Brinzeu" Emergency County Hospital Timisoara, Romania (No. 173/03.10.2019).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Work Stress, Mental Health and Validation of Professional Stress Scale (PSS) in an Italian-Speaking Teachers Sample**

**Pierpaolo Limone <sup>1</sup> , Roberto Zefferino <sup>2</sup> , Giusi Antonia Toto 1,\* and Gianfranco Tomei <sup>3</sup>**


**Abstract:** This study aimed validate the Italian version of the Professional Stress Scale (PSS). A questionnaire was translated into Italian and administered to two sample groups. The first group (*n* = 200) was the control group and the second (*n* = 1137) the experimental group. The participants in the study were students enrolled in a special needs training teacher course or a specialization course that aims to train support teachers. The study conducted two analyses; factor and reliability analyses. The factor analysis utilized the Kaiser-Meyer-Olkin (KMO) test which had a result of 0.925 for the scale; this was above the acceptable value of 0.7. The research studied 33 items and the BTS was significant for the 33 items scale (χ2 (528) = 4353.508, *p* < 0.001). Moreover, five eigenvalues greater than 1 were identified in the data, whereas the total variance explained was 63.7%. The reliability test utilized the Cronbach's Alpha score (0.936) of the scale and the value is calculated based on the response of 1106 individuals. The value is well above the value of 0.80, which indicates a high internal consistency level of the different items of the scale. This study showed that the Italian version of the PSS is a reliable and valid measure that can be used for research and clinical purposes.

**Keywords:** work stress; mental health; Italian professional stress scale

#### **1. Introduction**

Stress is often associated with negative experiences or notions of distress and negative effects that are related with the incapacity to deal with them. Every job can produce a certain level of stress which can affect individuals at all levels in the organization, from employees to managers to senior leaders. If stress is occasional it does not appear to be harmful, but problems often arise when it is chronic. Some of the sources of stress at work include interactions among employees, the workload, personal responsibility, and conflicts between home and work. Stress is known to cause adverse effects on health, physically and mentally [1,2]. For instance, it can lead to an escalated anxiety, substance addiction, burnout and depression. Employees who feel stressed at work are highly likely to get involved in unhealthy behaviors like poor diet, smoking, drug abuse, and alcoholism. Reports show that excess work stress leads to approximately 120,000 deaths and a total of approximately \$190 billion in health care costs annually [3] around the world. This is about 5–8% of the healthcare spending of national economies, which is a result of high direct expenditures (about \$48 billion), lack of insurance cover (about \$40 billion), and conflict (\$24 billion) [4]. Work stress can worsen an existing mental health problem, thus making it harder to control. Common mental and physical health issues and stress can exist independently, meaning people can be stressed about work and at the same time experience physical changes like high blood pressure with no depression or other mental health conditions. They may also have depression without experiencing stress. The effects

**Citation:** Limone, P.; Zefferino, R.; Toto, G.A.; Tomei, G. Work Stress, Mental Health and Validation of Professional Stress Scale (PSS) in an Italian-Speaking Teachers Sample. *Healthcare* **2021**, *9*, 1434. https:// doi.org/10.3390/healthcare9111434

Academic Editors: Alberto Modenese and Fabriziomaria Gobba

Received: 13 September 2021 Accepted: 23 October 2021 Published: 25 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

of work stress include reduced job satisfaction, absenteeism, poor delivery of services and higher mortality rates.

Employees today are experiencing greater challenges than ever because of company strategies, policies, restructuring, technological advancements, and deadlines that lead to high levels of professional stress. Professional stress occurs due to a discrepancy between the demands at the workplace and the capacity of an individual to meet these demands [5,6]. Professional stress greatly affects productivity, mental, physical, behavioral, and emotional health among professionals. The Professional Stress Scale (PSS) is a self-reporting technique for establishing sources of stress for professionals. It is utilized to determine the degree of stress experienced by professionals in the healthcare workplace (specifically referencing mental health professionals who work in schools such as school psychologists) and in the school environment [7,8]. This paper aims to discuss work stress, mental health and validate the Professional Stress Scale. For this purpose, the 33-item PSS scale, which includes four subscales, was administered to 200 individuals in a cool group and 1137 individuals in an experimental group. It had a Cronbach's Alpha score of 0.936 and KMO score of 0.925.

Most jobs may promote work related stress, particularly nowadays it involves the healthcare sector [9–11]. There are few studies carried out on mental health professionals, whose work is often considered less interesting, however, lots of research shows that these professionals may display high vulnerability to stress. Reports show that psychiatrists seem to have the highest rates of suicide among healthcare practitioners [12,13]. Other authors report that mental health professionals experience a wide range of problems which lead them to drop out of training [14]. According to [15], there is a perceived distress level of 59% of clinical psychology trainees, which is higher than in other groups. If the stresses of mental health professionals have to be mitigated, it is important to systematically analyze the stressors that cause them. Some studies show that mental health professionals face different stressors from other professionals in other healthcare sectors. Mental health studies show that besides being susceptible to stressors that are inbuilt in the workplace, mental health workers tend to face more challenges. For instance, a descriptive literature review reveals that the two primary sources of stress in healthcare facilities are patient contact, and organizational and administrative factors. According to the research conducted by Travers and Firth-Cozens [16], some of the causes of stress in the workplace include lack of resources and shortage of personnel, along with the major one being violent experiences with patients. While most of the problems in psychiatry field seem to be common to all health workers, the impacts may be aggravated by the intrinsic nature of the job. As such, Ref. [17,18] show that there are unique problems in this group that should be addressed. Thus, there is a great need for more research on the specific stressors faced by mental health workers. Hellman et al. [19] established five key stressors related to the therapeutic role of mental health workers in the public and private sectors. These factors include scheduling, professional uncertainty, personal degradation, work overload, and sustaining a therapeutic relationship. In a study of psychologists, Cushway and Tyler [20] noted that factors like self-doubt and client distress were stress factors. Additionally, other factors like organizational issues and workload were revealed. The coronavirus has changed the way people live and work. The study by Irawanto et al. [21] revealed that working from home, work-life balance, and work-related stress have significant effects, both direct and indirect, on job satisfaction. In fact, COVID-19 has changed working conditions due to social distancing policies. As many workers have begun to use new technologies at work, the study by Oksanen et al. [22] probed the potential effects of social media communication (SMC) stress on work. The results indicate a disparity in the resilience of workers during remote work and highlight the need for organisational support [23]. The study by Hong et al. [24], on the other hand, examined the associations between work overload, parental stress, work-family conflict, and job satisfaction during COVID-19. Seven hundred and eighteen kindergarten teachers participated in the study.

Labour resources can buffer the deleterious effect of adverse work environments. Stress reactivity can be an important work resource on a personal biological level. Deng et al. [25] examined how stress responsiveness interacts with work environments in predicting work burnout. Social stress at work appears to accelerate the loss of resources over consecutive working days. Elfering et al. [26] analyse workplace social stressors and resources possessed. The study by Wang et al. [27] aimed to investigate the prevalence of burnout as well as anxiety, depression and stress in resident physicians and evaluate the effects of an online psychological intervention on the mental health status of physicians with a high degree of burnout. The work-related well-being of an employee is to some extent related to the work environment perceived by colleagues rather than absolute [28,29]. It found that perceiving colleagues as having higher or lower demands than themselves is associated with lower job satisfaction and higher levels of emotional exhaustion. Therefore, the processes of social confrontation regarding job requests can influence the well-being of employees.

There are contexts such as education, services and the helping professions where nothing material is built and such immateriality of the work product can unsettle the individual. The lack of a material aspect makes, for some, the work objectionable because the results are not visible and immediate. In these areas, if the worker has not made strategic alliances, his or her work life can be very problematic. For this reason, in order to better evaluate the dynamics related to work-stress, the need to validate the PSS tool in Italian emerged. It also it seemed useful and interesting to validate the tool with educational professionals: individuals who often have difficulty in immediately recognizing the fruits of their labor.

In order to verify the presence of a work-related stress situation for these professionals, the Professional Stress Scale (PSS) which investigates through appropriate questions certain parameters such as workload, difficulty in working with others, lack of resources, conflicts with colleagues and superiors, self-esteem and home/work conflicts was administered. Although the standardized version has a good diffusion, in the specific Italian context the need arises for the validation of tools that make research in work-related stress a sector of greater depth [30–32]. Today's work with the use of technologies, multitasking, e-mails, the Internet insert today's worker into a "network" in which at times they risk being harnessed.

The present study thus aimed to validate the Italian version of the Mental Health Professionals Stress Scale, developed in 1996 by Cushway and colleagues. The Mental Health Professionals Stress Scale (MHPSS) measures stress for mental health professionals using the self-report method and identifying sources of stress. The expected relationships between the scale and between the criterion measures—the General Health Questionnaire, a symptom check list, job satisfaction, self-reported stress level and quality of social support—were demonstrated. The results also provide evidence for the discriminant validity of the subscales to measure different aspects of the stress experience [20].

The Italian version of MHPSS was used, prior to the present validation, in two studies by Zefferino and colleagues in 2006 [33] and 2009 [34].

The aim of the 2006 study was to identify the presence of sources of stress in an urban police emergency team and the causes of such stress using the PSS test and biomarkers such as salivary cortisol and interleukin 1β. In the 2009 study Zefferino investigated stress using a double approach: (i) a psycho-diagnostic test able to show psychological effects and (ii) a kit test able to measure salivary markers of stress as cortisol and interleukin 1β.

The present study responds to a pressing need to structure a protocol for the diagnosis and treatment of psychophysical stress within the services of the University Hospital of Foggia.

#### **2. Methods**

#### *2.1. Study Design*

Often, the subject does not know at the end of the day what his work consisted of. There are claims that in areas such as education, services and the aid professions nothing

material is built, and the immateriality of the product of work is something that can upset the individual, as the lack of a material aspect makes work critical. The Mental Health Professionals Stress Scale (MHPSS) is a self-assessment method for identifying sources of stress for mental health professionals. The 42-item scale, which includes seven subscales, was administered to 154 clinical psychologists and 111 mental health nurses. MHPSS was found to have good internal consistency (α = 0.87 for clinical psychologists; α = 0.94 for mental health nurses). Preliminary evidence suggests that the simultaneous validity of MHPSS is good. The modified mental health Professional Stress Scale (PSS) is used to assess self-perceived work-related stress [31]. This research included a validation of the Italian Version of Professional Stress Scale [32,33]. The questionnaire consists of 33 questions and detects the frequency of occurrence of certain stressful events using the following 4-value scale: 0 = never happened, 1 = doesn't happen usually, 2 = happens occasionally, and 3 = happens to me.

Emotions, thoughts, workload, relationships with colleagues and superiors, and the ability to reconcile professional and personal life are investigated. The total score is equal to the sum of the scores obtained for each item. Higher values indicate a greater degree of work-related stress.The items and the answer alternatives are easy to understand. In two previous experiments, the questionnaire was applied to a specific category of workers (an urban police team) who, by answering a specific question, showed maximum understanding of the questions. Furthermore, the questions are of a general nature and, therefore, are free from content specific to any subpopulation. To examine the proposed PSS, Zefferino et al. [34] conducted an explorative factor analysis (EFA) and a confirmatory factor analysis (CFA). The authors reported that there was support for validity of the Professional Stress Scale.

These results showed that the factors present an acceptable validity and reliability. In addition, the instrument was shown to have adequate convergent validity with theoretically related constructs. All constructs exhibited composite trait reliability levels that exceeded 0.7 [34], ranging between 0.87 and 0.95.

#### *2.2. Sample*

Subjects participating in our study were teachers enrolled in a TFA support specialization course, a course that aims to train support teachers. The sample was selected in order to investigate the level of perceived stress in teachers and validate the scale on this sample. The authors and research team evaluated the intentional sample through the technique of heterogeneous sampling. This type of sampling is intended to provide a wide range of cases relevant to a particular phenomenon or event. The purpose of this type of sample design is to provide as much information as possible about the event or phenomenon under consideration. In the case of the present research, it was deemed useful to analyze the sample of teachers enrolled in the TFA support specialization course.

The study involved samples from two groups, a control group (*n* = 200), and an experimental group (*n* = 1137), but the results are based on the information obtained from the responses of 1106 respondents.

An attempt was made to make the experimental group as large as possible to reduce the influence of non-obvious differences between the subjects, and the aim was also to simultaneously reduce the probability of incurring a first and second type of error.

To build the Italian version of a foreign instrument, it is necessary to start with data collection to verify the validity and reliability of the tool for the Italian context. Once these have been verified, the standardization sample can be collected. The standardization sample should normally be proportional to the population at which the test is aimed and present samples similar in characteristics to those present in the original manual.

The control group was made up of 200 people who represent in percentages the same demographic and socio-cultural characteristics (educational qualification, geographical origin) of the experimental group (25% belonging to each grade of kindergarten, primary, secondary, and secondary school second degree and maintaining the same gender ratio).

The pilot study is developed in the Italian context, where the initial training course for teachers is online and distributed nationally; therefore, teacher interviews from all areas of Italy (67% from southern Italy) were included. By pilot study, we mean it is an exploratory test survey to demonstrate that the Italian version of the PSS test can work so that it can subsequently be used on a large scale. It is therefore presented as a feasibility study intended to test the questionnaire to guide the application of the test on large dimensions.

The Italian version of the questionnaire was administered to teachers on the initial training course to support teachers at the University of Foggia (*n* = 1106). The data were broken down by demographic profile, response processing and educational level. The data was provided via a Google form in December 2020 during the COVID-19 state of emergency. Using the online form made it possible to receive results in real time and to quickly view a summary.

#### *2.3. Data Collection Instrument*

The research utilized questionnaires to collect information and the professional stress scale to determine the factors contributing to stress in the workplace.

#### *2.4. Ethics*

The research study complied with the general ethical principles of the Declaration of Helsinki and was approved by the research team's University Institutional Review Board, protocol code 40979-III.11 and approved on 6 August 2021 issued by La Sapienza University of Foggia.

#### *2.5. Statistical Analysis*

The research performed a t-test to assess the differences between the means. A chi square test was also conducted to establish the statistical significance of the subscales. To evaluate the external validity, the research made comparisons using the PSS, to determine the level which occurrences are considered stressful. The PSS version that was used was a 33-item tool. The research utilized the Cronbach's Alpha score, which was 0.936, to determine consistency. The study also conducted a factor analysis using the KMO test which was 0.925.

To establish if the collected information was suitable for the analysis, we applied the Kaiser-MeyerOlkin (KMO) measure of sampling adequacy and BTS tests. The KMO value was acceptable (0.943) and the BTS was significant (χ2 (528) = 18,361.702, *p* < 0.001). As such, the data was relevant for factor analysis; thus, a principal components analysis was performed. The results showed that the majority of the items showed the factor loading values of 0.5 and above, whereas items 6, 29, 31, and 32 have factor loading values of 4 or above, which are acceptable values in research. Since the chi-square value is less than 3 (2.360), it shows that the model was adequate.

#### **3. Results**

#### *3.1. Sample Description*

Data was taken from Google modules in reference to a sample of 1137 people. The experimental group consisted of school teachers enrolled in the TFA support course as students. The subjects belong to the 5th cycle of the University of Foggia, of which 85.7% are women and 14.3% men. Although there are only slight numerical differences with respect to the degree of origin, the most representative degree was lower secondary school (29% of the participants). The age varied between 20 and 60 years.

A total of 1106 respondents completed the questionnaire. The 33 items were evaluated through the PCA extraction method in which the majority of the items had factor loading values of 0.5 and above, while a few had 4. The KMO test result value was 0.943, an acceptable value of above 0.7. The BTS was significant for the 33 items scale (χ 2 (528) = 18,361.702, *p* < 0.001). Also, 6 eigenvalues greater than 1 were identified in the data, while the total

variance explained was 60.8%. Internal consistency Cronbach's a for the entire sample was found to be 0.936.

#### *3.2. Reliability Analysis*

Table 1 shows the Cronbach's Alpha score (0.936) of the scale and the value is calculated based on the response of 1106 individuals. The value is well above the baseline value of 0.80, which indicates a high internal consistency level of the different items of the scale. Moreover, Table 2 shows the item statistics that include the values of the mean and standard deviation of all the items of the scale.

**Table 1.** Cronbach's Alpha.



#### **Table 2.** Item Statistics.

*3.3. Factor Analysis*

The 33 items of the scale were examined through the PCA extraction method. As shown in Table 3, the majority of the items showed factor loading values of 0.5 and above, whereas items 6, 29, 31, and 32 have factor loading values of 4 or above, which are accepted values in research [35,36]. The demographic items were not included in the factor analysis.


**Table 3.** Eigen Values, KMO & BTS Tests.

Extraction Method: Principal Component Analysis

The KMO test result value was 0.943 for the scale, which is also well above the acceptable value of 0.7. The BTS was also significant for the 33 items scale (χ 2 (528) = 18,361.702, *p* < 0.001) (Table 3). Moreover, six eigenvalues greater than 1 were identified in the data, whereas the total variance explained was 60.8%. Besides, a scree plot (Figure 1) also showed the six identified eigenvalues of 11.1, 2.8, 2.4, 1.4, 1.3, and 1.2.

**Figure 1.** Screen Plot.

#### **4. Discussion**

To verify the presence of a situation of work-related stress, the PSS was administered, which investigates through appropriate questions some parameters such as workload, difficulty in working with others, lack of resources, conflicts with colleagues and superiors, self-esteem and home/work conflicts. The effectiveness of these scales for measuring stress has been studied in various mental health settings, ranging from nurses, to university students, to public administration workers [37,38]. Edwards and Burnard [39] revealed an excessive level of workplace stress for mental health nurses. The most frequently reported sources of stress were administrative and organizational concerns, patient issues, heavy workload, interprofessional conflict, financial and resource issues, professional self-doubt, home/work conflict, staffing levels, changes in the health service, maintaining standards, lecturing and teaching, long waiting lists, and poor supervision.

Lee et al. [40] identified, by analyzing a sample characterized by psychologists, nurses, and social workers that for depression and anxiety, that scores appeared slightly different across professional groups. In fact, nurses and social workers showed significantly higher total scores than clinical psychologists, and there were significant differences in subscale scores among professionals. Results and reviews conducted to date have suggested that the scale was a useful measure and predictor of stress [41]. Specifically, in this study, the

PSS was administered to professionals in the educational setting; indeed, it was found that there are contexts such as education, service, and helping professions in which the immateriality of the work product can upset the individual. For this reason, in order to better assess the dynamics related to work stress, the need emerges to validate the PSS tool in Italian.

Challenges, beyond the dimensions already highlighted, could also lead to depression, high blood pressure, and fatal coronary conditions [42]. As such, it is vital to conduct the occupational stress scale because it determines the magnitude of stress experienced by professionals in the healthcare workplace. The results of this study show that the PSS has good characteristics and can be used in future research.

#### **5. Limitations of the Study**

The data provided in this study do not represent the whole Italian employed population, as the study focused on mostly teachers rather than on workers in all sectors of the economy. The Italian healthcare setting comprises of many mental health professionals; thus, a more representative sample should be examined for future studies. Although the sample is not broadly representative of different work contexts, we believe it can be used in other work contexts as well since the sample was large (*n* = 1.106). The non-random sampling technique used to recruit the sample could have contributed to selection bias and lack of representativeness.

Thus, future research should consider sampling workers from all sectors to attain more extensive results. Lastly, the study does not categorize and discuss individual subscales; thus, the results are linked to only two classes of workers (police in the experimental phase and teachers in this validation), and they should also be extended to other categories to generalize the clinical results in the Italian context. Therefore, future observations should categorize the subscales accordingly for easy analysis and interpretation of the results.

#### **6. Conclusions**

The results of this research indicate that the Italian version of the Professional Stress Scale has excellent features, and thus, it qualifies to be used for future research. The initial evidence concerning the usage of the PSS is quite intriguing and inspiring, even though there is still a need for more information and analysis. Thus, reliability analysis and more validation of the PSS are worthwhile objectives for future studies. Besides providing evidence of the validity of PSS with other professionals, a study like this could improve the understanding of the causes of stress for professionals in the workplace.

**Author Contributions:** Conceptualization, P.L. and R.Z.; methodology, G.A.T.; validation, G.T. and G.A.T.; formal analysis, G.T.; writing—original draft preparation, G.A.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of University of Foggia (protocol code 40979-III.11 and approved on 6 August 2021).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

