Connotations of psychological and physiological health in the psychosocial work environment: An industrial context

3.1 Participants

The study participants were technical workers working in operational, maintenance and production activities in the petrochemical industries of Peninsular Malaysia. Initially, 300 total responses were collected, but due to missing values, the responses of 277 subjects were used in the study. Of these responses, 210 were male and 67 were female aged between 20 and 49 years. All the participants were healthy. Exclusion criteria included the use of illicit drugs, use of any prescribed medication, physical activity practice of more than five consecutive hours without having a leisure break of not less than thirty minutes of duration, provided 8×5 hours per week. Other conditions, especially for women such as pregnant or having any disturbance in regular menstrual cycles and ovulation, were also the part of exclusion criteria.

3.2 Sampling design

Multi-stage sampling was used due to large inquiries extending to the considerable large geographical area. The first stage in multi-stage sampling is to select the large primary sampling units like states, then areas and finally people within the selected areas [15]. Javaid et al. in their study have proposed the multi-stage sampling procedure for the petrochemical industries of Malaysia, which this study followed [16].

The study constructs were adapted from the second version of Copenhagen psychosocial questionnaire COPSOQ II [18], a thorough questionnaire that covers all the aspects that are important to study the psychosocial work environment along with health and wellbeing [19, 20].

3.4 Procedure and measurements

For data collection local enumerators were hired and given a full workshop to understand the questionnaire survey along with the other study protocols. To reduce any potential bias, careful monitoring was done by the authors.

The BMI is a non-invasive health biomarker, which was calculated based upon every subject’s height and weight. The recommended protocols for calculating BMI (height and weight) were fulfilled [22]. A participant’s one-time height was calculated with the help of the Seca-213 mobile height stadiometer, and weight was computed three times with the help of the Omron HBF-375 digital weight machine. The average weight was taken out for the evaluation of the BMI.

The latest second generation technique structural equation modeling allows researchers particularly in health to examine and analyze complex and causal relationships in explaining the development of the phenomena such as diseases and health behaviors. The biggest advantage of using SEM in epidemiology studies over first generation research techniques such as multiple regressions is that it is more powerful and can manage measurement error and the path coefficients are measured simultaneously. Therefore the SEM method, Partial Least Square (PLS) was implemented using SmartPLS 3.2.7 [30] for data analysis.

Table 2 shows the demographic characteristics of the participating respondents who were technical workers of petrochemical industries of Malaysia. The character traits included were the gender, age, education, and experience of the respondents. The percentage of male respondents was high due to large male-dominated representation in petrochemical plants [31]. This was seen at the time of data collection in that 75.8% were male while the remaining 24.2% were females of the total study sample. Further details of the character traits of the participants are mentioned in the Table 2.

4.1 The measurement model

Table 3 shows the results of measurement model. The convergent validity was established through item loadings, composite reliability and the average variance extracted. The item loadings of all the constructs were above the required threshold value of 0.7 [32] except for two items, but, as recommended by [32] they were only considered for removal if the composite reliability (CR) and AVE of that variable were below the acceptable range. As seen in the table, the CR and AVE of all variables were greater than the suggested threshold value of 0.7 for CR and 0.5 for AVE [33].

The discriminant validity was assessed through the Fornell and Larcker criterion [34], which requires that the square root of AVE for all the variables should be greater than correlations between all the variables. As shown in Table 4, all the values on the diagonal are greater than the correlations between the variables ascertaining discriminant validity. However, a recent concern has been observed about the Fornell and Larcker criterion. Henseler et al. proposed an alternative and more robust approach for establishing discriminant validity, i.e., the Heterotrait-Monotrait Ratio (HTMT) [35]. In Table 5, all the HTMT values are given, and all the values are comfortably below the conservative threshold value of 0.90 [35]. Thus, both criteria confirm the discriminant validity of all the constructs.

After achieving the measurement model, the next step is to examine the structural model, which is also called as the path model. It is recommended to report R², beta, and the t-value, using the bootstrapping technique of 5,000 resamples [33]. In addition to that, Kaufmann and Gaeckler suggested reporting Q² values (predictive relevance) [36]. Tables 6 and 7 present the results for the direct relationships, and considerable support exists for hypotheses H₁₁ to H₁₆ but no statistical support was present for hypothesis H₁₇. More interesting results exist for the direct relationships between WFC, QD, ED, and JI with BMI. The support for these hypotheses indicates that these variables has a more direct and impactful role in improving BMI. ED and WFC also has a direct relationship with BO; however, no support was found for hypotheses H₁, H₃, H₄, and H₈.

Table 8 presents the results of the structural model, and the results show support for four of the ten proposed hypotheses. The results show statistical support for the mediation path for H₁₉ (β= 0.166, t = 3.458), H₂₃ (β= 0.083, t = 1.966), H₂₅ (β=–0.121, t = 3.655) and H₂₇ (β= 0.212, t = 4.583). Remaining hypotheses showed no statistical support and were rejected. Thus, support was found for the mediating role of stress for emotional demands, quantitative demands, the social support of colleagues, work-family conflict, and burnout.

Table 9 shows the summarized results of predictive power. According to Cohen (1988), the R² value of both stress and burnout was large (>0.35), and predictive relevance (Q²) was achieved for stress and burnout (>0), while no predictive relevance was found for the BMI.

5.1 Enlightenment on direct and indirect effects

All behavioral demand factors such as quantitative demands, emotional demands, work-family conflict and job insecurity had a significant and strong direct association with the BMI of the workers except for one behavioral resource factor, i.e., social support of colleagues. The direct relationship between demands (WFC, QD, ED, and JI) and BMI indicates the possible role of context [37]. Wu et al. found that no significant relationship existed between physiological factors and social support even after adjusting for age, education, and income. They have argued that to control the biomarkers of health such as BMI in this case other pathophysiology needs to be examined besides social support [38].

Job resources vary their roles based on prevailing job characteristics. Besides, social support of colleagues, many other coping factors that are both action-oriented and intra-psychic exist that need to be managed, and these psychosocial factors generally depend based on an individual’s perception of the resources available. There are characteristics in the working environment that some workers would perceive as resources while others would not. For instance, petrochemical industry workers may be happy when they feel that the assigned tasks are manageable and within their control at one point in time. However, in the next hour, a worker’s positive emotions can be disturbed due to situational factors in which supervisory support is more influential than is the support of colleagues. This within-person fluctuation because of changing behaviors may also result in psychological and physiological changes.

However, limited literature has focused on the role (quality) of leadership in ensuring less obese and healthy employees, even though the quality of leadership is key to the success of any policy or practice [39]. The success of any health-related programs or policies introduced or implemented by an organization are largely dependent on the active involvement of their leadership. Especially with the advent of health-specific leadership HS-L [40] a leadership style, which primarily focuses on the improvement of health and ensuring that any health-related issue of employees is dealt with before it becomes chronic, is critical.

The psychological conditions of workers may also affect cultural values. When work is the central feature of life, sources of conflict outside work such as family domain may cause workers to expend efforts and time in an area upon which they are not as focused, diminishing health and leading to conditions such as burnout syndrome. Accordingly, double engagement in work and family may lead to burdensome experiences and be less beneficial. Past studies have acknowledged that Malaysia is a collective society in which harmony is desired and any kind of confrontation is seen as a threat, which is consistent with studies in which authors have found that occupational stress factors created a highly stressful situation in Malaysian workers [41].

Emotional Demands (ED) was the only variable that had a significant and direct effect on both the psychological and physiological health of the workers. Past studies [42] have shown that emotional energy is a main distraction influencing psychological availability. Emotional energy demands that workers exhibit personal expressions in tasks that require a certain level of emotionality, which engages and influences psychological availability. Emotional Demands in many studies has been found to be related to emotional labor wherein workers look for emotional resources to meet demands and minimize stress. If the provided resources are not sufficient enough, then emotional labor charged with emotional demands results in an impairment of health that further exhausts the mental and physical resources of workers, which, in turn, leads to the depletion of energy, i.e., emotionally exhausted workers with health problems as shown by the study sample.

Past researchers have also acknowledged the connection between workplace stress and mental health of workers in the oil and gas industry wherein psychosocial factors affect the psychological health state [43]. In the oil and gas industry, psychosocial risks can cause ill-health for individuals or groups due to long-time exposure to poor working conditions [44]. The results also align with the findings of the past researchers, which found that stress was strongly correlated with the adverse working conditions of the petrochemical industry [45].

Insecurity creates anxiety and stress and distracts workers from their work. Psychological availability also corresponded with how secure workers felt about their work and status. Job insecurity is surrounded by dimensions like a lack of self-confidence, heightened self-consciousness, and uncertainty about the fit with the industry and its purposes.

A lack of self-confidence is a particular issue for new or low-status workers. The demographic characteristics of this study have shown that almost 63% of the workers are those who are new to work and have less than one year of experience within the existing organization. The operational activities in petrochemical plants need a high level of concentration and motivation due to excessive petrochemical products production. A new worker can show anxiousness about how the speed and quality of his work are compared to a coworker in the plant. The psychological factors keep coming up, forcing him to think of questions like “Was I working fast enough or was I doing the work correctly enough? Are the employers going to keep me or terminate me because of my small contributions?”. So, at times workers tend to worry more about performed tasks and the way that these tasks will be perceived by the employer rather than considering the actual task itself. This lack of self-confidence is a product of psychological factors that are firmly associated with job insecurity and lead to stress experienced by the workers.

Heightened self-consciousness arises when workers are focused on how others perceived and judged them (whether or not such judgment actually occurred). This type of psychological thinking forced them to be distracted rather than to be engaged with assigned tasks. This dimension of job insecurity normally arises when workers perceive themselves, consciously or not, as stage actors surrounded by audience and critics, rather than as co-workers who simply do their jobs. A self-consciousness worker engages himself in managing impressions rather than focusing on the task at hand.

Fit uncertainty preoccupies workers, leaving them little space, energy or the desire to express them at the time of task performance. Such a psychological condition pulls a worker’s concentration from tasks and forces him to show a lack of commitment. A worker’s uncertainty about his fit within the industry and its purposes make him struggle with his desires to contribute to achieve the end goals of his systems because of a low willingness and often leads to absenteeism due to sickness.

The demographic characteristic of this study showed that around 31% of the respondents fell in the category of low-qualified with only a primary education and around 24% had only a secondary education. Therefore, based on past studies and the results of the current study arguments can be presented that a reason for the unhealthy behaviors and increased health impairments of respondents in this study was the increased psychological stress that workers faced in the forms of psychosocial job demands, emotional demands, job insecurity and work-life conflicts. Additionally, the European working conditions survey showed that low-qualified workers receive low social support in comparison to more qualified workers. Low-qualified workers not only receive low social support but also tend to have less access to occupational training for professional development and health promotion. Therefore, the combination of low social support and high psychosocial demands are both associated with increased health risk.

This current study has significantly contributed to the determination of a difference between stress and burnout. This study found that stress significantly contributed as a mediation variable for the outcome of the psychological health variable of burnout. The study results have shown that stress and burnout seem to have a few different antecedents, correlations and outcomes as discussed in the past studies [46]. This study found a significant and direct relationship of all the behavioral factors with stress; however, only emotional demands and work-family conflict significantly affected the burnout of members of the study sample which is in lieu to studies where psychological demands are significantly associated with the burnout [47]. Interestingly, the burnout concept is often conceptualized and overlapped with the framework of stress. This study also found that the mediating role of stress is significant in quantitative demands, emotional demands, work-family conflict and social support of colleagues’ relationship with burnout. The study results clearly demonstrate that stress and burnout are different concepts, and stress leads to burnout.

The limitation of this study was the time frame and availability of the workers to be part of longitudinal study. Future studies should objectively measure health biomarkers by conducting longitudinal studies and to calculate the worker’s biomarkers at two different points of time by adding different physiological and physical biomarkers. This may help future research studies to determine the causal relationships more strongly. Future studies should consider the role of quality of leadership as a job resource factor, in particular, the role of Health-Specific Leadership (H-SL) which is a kind of leadership that can show commitment and dedication for ensuring employees health. This type of leadership can convey to the employees an organization’s seriousness about their health. Therefore, the study proposes that health-specific leadership be considered as a key factor in future studies with the BMI issue of employees. Employee’s commitment and grit is required to become a fitter and for that purpose HS-L can play an important role in conveying organization resolve for a fit workforce and solidifying employee’s dedication for good physiological health.