Influence of Parenting Quality and Neuroticism on Perceived Job Stressors and Psychological and Physical Stress Response in Adult Workers from the Community

Introduction

Job stress is caused by jobs that are not matched to a person’s knowledge and abilities, involve pointless tasks, unpleasant duties, too much or too little work load, lack of interaction with colleagues, harassment, lack of communication, and conflicts between family life and work.¹ Job stress is associated with depressive symptoms.² Burnout syndrome resulting from job stress is associated with musculoskeletal and cardiovascular diseases, and reduced life expectancy.^3,4 In 1979, the job demand-control model, which shows that the combination of high job demand and low decision latitude is associated with high psychological distress, was proposed.⁵ Later, the job demand-control model was extended to the job demand-control support model to account for the role of workplace support as a third predictor.^6,7 In 1988, the job stress model was established by the National Institute for Occupational Safety and Health (NIOSH), which showed that age, sex, marital status, personality traits, etc., affect acute reactions by job stressors. Based on this job stress model of the NIOSH, the Generic Job Stress Questionnaire (GJSQ), which is a self-completed job stress questionnaire, was developed after appropriate items (eg, physical environment) were selected.⁸ In Japan, job stress is also considered as a major risk factor for a wide range of health outcomes.⁹ In 1993, the Japanese version of the GJSQ was translated and confirmed to be sufficiently reliable and valid.¹⁰ The Brief Job Stress Questionnaire (BJSQ), which contains components similar to GJSQ, was developed referring to the job demand-control support model,⁵ and is used in the stress check program in Japan.^11–14 Following the amendment of the Industrial Safety and Health Law in 2014, the BJSQ has been used as the standard questionnaire in the stress check program, which has been compulsory since December 1, 2015, for business establishments with 50 or more workers.⁹

Personality traits, particularly neuroticism, have been reported to be associated with depressive symptoms in the general adult population.^15–18 The association between neuroticism and job stress has also been reported. In recent years, studies have reported that neuroticism is more strongly associated with job stressors and burnout syndrome caused by job stressors than work-associated factors.^19,20 These studies do not support the widespread idea of previous researchers that the psychosomatic stress response is primarily linked to work-contextualized factors and is only secondary to personality dispositional factors.^20–22

Parenting styles are the child-rearing that children receive from their parents, which reflect the “bonding” between parents and children, and are closely associated with child development and the onset of mental illness in adulthood.²³ Poor parenting styles and child abuse in childhood have been reported to affect depressive symptoms in adulthood.^24–27 The association between parenting styles and neuroticism has also been reported. Certain personality traits (extraversion and neuroticism) have been shown to mediate the effects of parenting styles on depressive symptoms in the general adult population.^15,18 In association with personality dispositional factors other than neuroticism, we reported that childhood abuse affects depressive symptoms in adulthood through affective temperaments (depressive, cyclothymic, irritable, and anxious).²⁸ In addition, child abuse reportedly affects depressive symptoms in adulthood through specific attachment styles.²⁹

There have been no studies to date showing the involvement of personality traits in the effects of parenting styles on job stressors and the stress response. Our recent studies reported that neuroticism mediates the effects of parenting styles and child abuse on the negative appraisal of adult life events and on adult depressive symptoms.^17,18 These findings suggest that parenting styles may also affect job stressors and the resulting stress response through their effects on neuroticism. We hypothesized that neuroticism acts as a mediator of the effects of parenting quality on perceived job stressors and the psychological and physical stress response (PPSR), and verified this hypothesis in adult workers using structural equation modeling.

Subjects and Methods

Subjects

Between April 2017 and April 2018, questionnaires and instructions of this study were distributed to 1,237 adults by convenience sampling through our acquaintances at Tokyo Medical University. A total of 640 subjects did not participate in this study, and 69 subjects were excluded due to missing data or nonworkers. A total of 528 subjects (233 men and 295 women, average age: 41.4 ± 11.9 years) who provided consent and fully valid responses were analyzed (Table 1). This study was part of a larger study, in which several questionnaires were investigated.³⁰ In terms of occupational contracts, there were 452 regular employees, 6 contract or commission employees, 52 temporary or part-time employees, and 15 others (Table 2). Regarding job class, there were 381 general class employees, 65 sectional manager class employees, and 7 manager level or higher employees (Table 2). Employment status was 323 inflexible, 24 discretionary, 5 flexible, 26 shift work (not including night shift), 103 shift work (including night shift), and 2 others (Table 2). Overtime work (hours per month) was 20 hours or less in 351 workers, 21–40 hours in 74 workers, 41–60 hours for 29 workers, and 61 hours or more in 16 workers (Table 2). Demographic information and the 3 questionnaires were anonymously surveyed. Participation in this study was not mandatory, and subjects were informed that participation was completely voluntary, that there would be no penalty for not participating, and that the collected information would be anonymized so that the individuals cannot be identified. This study was approved by the Ethics Committee of Tokyo Medical University.

Table 1 Patient Characteristics, PBI, EPQ-R, and BJSQ Scores and Their Correlation with PPSR Scores of BJSQ or Effects on PPSR Score in Adult Workers

Table 2 Effects of Occupational Contracts, Job Class, Employment Status, and Overtime Work Hours on PPSR Score of the BJSQ in Adult Workers

Questionnaires

Parental Bonding Instrument (PBI)

PBI is a retrospective self-administered questionnaire that measures how respondents evaluate their parents’ child-rearing, after they become adults.²³ The PBI consists of 25 question items, which are classified into 2 subscales; 13 question items of overprotection (0–39 points) and 12 question items of care (0–36 points), and it is designed to be evaluated on a 4-point scale of “very likely (3 points)”, “moderately likely (2 points)”, “moderately unlikely (1 point)”, and “very unlikely (0 points)”. The PBI is performed separately for each father and mother. The total scores of overprotection and care of the father and the mother were used for analysis. The validity and reliability of the Japanese version of the PBI were confirmed.³¹

Eysenck Personality Questionnaire-Revised (EPQ-R)

The Eysenck Personality Questionnaire is a self-administered questionnaire that measures neuroticism, and was revised in 1985 to the Eysenck Personality Questionnaire-Revised (EPQ-R).³² The shortened version of the EPQ-R consists of 12 question items, and it is designed to be evaluated on a 2-point scale of “yes (1 point)” and “no (0 points)”. The total score was used for analysis. The validity and reliability of the Japanese shortened version of the EPQ-R were confirmed by our previous study.³³

Brief Job Stress Questionnaire (BJSQ)

The BJSQ is a self-administered questionnaire that assesses job stress, which was developed by Shimomitsu et al.^11,14 The BJSQ consists of 57 question items, which are classified into the following 4 subscales: perceived job stressors, the PPSR, social support, and satisfaction, and it is designed to be evaluated on a 4-point scale (1 point to 4 points). In this study, 2 subscales of 17 question items, ie, perceived job stressors (17–68 points), such as quantitative job overload, qualitative job overload, physical demands, job control, skill utilization, interpersonal conflict, poor physical environment, suitable job, and meaningfulness of work, and 29 question items of the PPSR (29–116 points), such as lack of vigor, irritability, fatigue, anxiety, depression, and somatic responses, were evaluated and analyzed. The higher the score of each subscale, the higher the stress. The total scores of perceived job stressors and the PPSR were used for analysis. In the BJSQ, as evaluation of job stressors is a subjective evaluation, job stressors are described as “perceived” job stressors in this study.

Structural Equation Model

In this study, neuroticism was hypothesized as having a mediating role on the effects of parenting quality on perceived job stressors and the PPSR. Based on this assumption, a structural equation model was built using care or overprotection scores of the PBI, the total score of the EPQ-R, and perceived job stressors and PPSR scores of the BJSQ. The average scores of each subscale of the father and mother as observed variables in the PBI composed a latent variable of care or overprotection (Figure 1).

Figure 1 The results of direct (A and C) and indirect (B and D) effects of covariance structure analysis in the structural equation model with parental overprotection (A and B) or parental care (C and D), neuroticism, job stressors, and PPSR in 528 adult workers from the community. Ovals show latent variables and rectangles show observed variables. The solid arrows indicate the statistically significant pathways, and the dotted arrows (A and C) indicate the nonsignificant pathways. The numbers beside the arrows represent the direct (A and C) or indirect (B and D) standardized coefficients. *p < 0.05, **p < 0.01, ***p < 0.001.

Statistical Analysis

For a comparison of demographic information and questionnaire data, the Pearson correlation coefficient was calculated, and the t-test or analysis of variance followed by the Bonferroni test was performed using SPSS Statistics Version 25 software (SPSS Inc., Chicago, IL, USA). We analyzed the structural equation model using covariance structure analysis with the robust maximum likelihood estimation method (Mplus version 8.0, Muthén & Muthén, Los Angeles, CA, USA). In this study, the Comparative Fit Index (CFI) and Root Mean Square Error Approximation (RMSEA) were used as fit indices. A CFI of greater than 0.95 and a RMSEA of less than 0.08 were considered to indicate an acceptable fit, whereas a CFI of greater than 0.97 and a RMSEA of less than 0.05 were considered to indicate a good fit in this study.³⁴ All coefficients of the covariance structure analysis were standardized.

Results

Correlation of Basic Demographic Information of the Subjects and Data of Each Questionnaire with PPSR Scores of the BJSQ

Table 1 shows the demographic characteristics and the scores of PBI, EPQ-R, and the BJSQ, and their correlation with PPSR scores of the BJSQ or effects on the PPSR score in 528 adult workers. PPSR scores were higher in women and unmarried individuals. No other demographic data was associated with PPSR scores. Paternal care and maternal care were negatively correlated with PPSR scores; the higher the care in childhood, the lower the PPSR scores. Paternal overprotection and maternal overprotection were positively correlated with PPSR scores; the higher the childhood overprotection, the higher the PPSR scores. The EPQ-R scores were positively correlated with PPSR scores; the higher the neuroticism, the higher the PPSR scores. Perceived job stressors were moderately positively correlated with PPSR scores; the higher the perceived job stressor scores, the higher the PPSR scores.

Cronbach’s α coefficients calculated for the individual subscales of the PBI were found to be 0.866 for overprotection from father, 0.920 for care from father, 0.873 for overprotection from mother, and 0.915 for care from mother and indicate excellent internal consistency. Cronbach’s α coefficient calculated for the EPQ-R was 0.859, indicating excellent internal consistency. Cronbach’s α calculated for the individual subscales of the BJSQ was found to be 0.740 for perceived job stressors and 0.909 for the PPSR and the latter indicates excellent internal consistency.

Table 2 shows the effects of occupational contracts, job class, employment status, and overtime work hours per month on PPSR scores of the BJSQ in 528 adult workers. Multiple comparison by the Bonferroni test of the factors that were statistically significant on ANOVA demonstrated that workers performing discretionary work had a significantly lower PPSR score than those performing inflexible and shift work (not including night-shift work). The association of other variables with the PPSR was not significant.

Analysis of the Structural Equation Model

In the model of Figure 1A, the fit indices were CFI = 0.999 and RMSEA = 0.021, indicating a good fit. The standardized coefficients from latent variable “overprotection” to observed variables were similar; 0.767 for fathers and 0.798 for mothers. Overprotection had significant direct effects on neuroticism and the PPSR, but not on perceived job stressors. Neuroticism showed significantly positive direct effects on perceived job stressors and the PPSR. Perceived job stressors had a significantly positive direct effect on the PPSR.

Figure 1B shows the significant indirect effects of overprotection. Overprotection had significant indirect effects on perceived job stressors and the PPSR via neuroticism. Furthermore, high overprotection was significantly associated with the high PPSR via combined pathways involving both neuroticism and perceived job stressors. The indirect effect of overprotection on the PPSR via perceived job stressors was not significant (data not shown in Figure 1B, standardized path effect = 0.071, p = 0.086). High neuroticism was also associated with the high PPSR through perceived job stressors. The square of the multiple correlation coefficient of PPSR scores in this model was 0.396, ie, this model explains 39.6% of the variability in the PPSR of adult workers.

In the model shown in Figure 1C, the fit indices were CFI = 1.000 and RMSEA = 0.000, indicating a good fit. The standardized coefficients from the latent variable “care” to the observed variables were 0.690 for the father and 0.783 for the mother; ie, the contribution of maternal care was larger than that of paternal care. Care had significant negative direct effects on neuroticism and the PPSR, but not on perceived job stressors. Neuroticism showed significant positive direct effects on perceived job stressors and the PPSR. Perceived job stressors had a significant positive direct effect on the PPSR.

Figure 1D shows significant indirect effects on care. Care had significant negative indirect effects on perceived job stressors and the PPSR via neuroticism. Furthermore, high care was significantly associated with low PPSR via combined pathways involving both neuroticism and perceived job stressors. The indirect effect of care on the PPSR via perceived job stressors was not significant (data not shown in Figure 1D, −0.018, p = 0.283). High neuroticism was associated with the high PPSR through perceived job stressors. The square of the multiple correlation coefficient of PPSR scores in this model was 0.386, ie, this model explains 38.6% of the variability in the PPSR of adult workers.

Discussion

To our knowledge, this study is the first report to suggest that the quality of parenting is indirectly associated with PPSR through effects on neuroticism and perceived job stressors, by performing covariance structure analysis on adult workers. Previous studies have reported associations between the quality of parenting and neuroticism and between neuroticism and job stress consequences.^15,18-20 However, how the quality of parenting is associated with perceived job stressors and the PPSR through neuroticism has not been reported to date.

In a previous cross-sectional study, overprotection from fathers was found to indirectly affect depressive symptoms through neuroticism in male patients with depression.¹⁵ However, there was no mediating effects of neuroticism in women. On the other hand, in a previous study of general adult volunteers, including both men and women, parenting styles influenced the negative appraisal of adult life events and adult depressive symptoms via neuroticism.¹⁸ The results of these previous studies are consistent with the results of the present study, which suggest that high parental overprotection is associated with high-perceived job stressors and the high PPSR through neuroticism, and parental care has the opposite effect. Therefore, our present results support the hypothesis that poor child-rearing received from parents first enhances neuroticism and, after a long latent period, increases the sensitivity to stress and at the same time enhances the stress response. Conversely, behaviorally inhibited children may influence poor parenting styles (eg, overprotection) from parents.³⁵ Considering our results, there is a possibility that poor child-rearing and children’s neuroticism make a vicious cycle.

In the structural equation model of this study, perceived job stressors had a significant effect (0.266:1A, 0.273:1C) on the PPSR, but neuroticism had a stronger effect on the PPSR (0.449: 1A, 0.460:1C) than perceived job stressors. The role of neuroticism has been reported in previous studies on job stress. For example, in the study of Lachowska et al, neuroticism had a greater effect on the symptoms of somatic disorders, anxiety, insomnia, functioning disorders, symptoms of depression, and global distress than work demands in working men or women raising children of up to 12 years of age.¹⁹ In a study by Bianchi et al using burnout at work as an index, neuroticism had a stronger effect on burnout than job stressors, which were evaluated by effort-reward imbalance.²⁰ These previous studies are consistent with the results of our present study. However, these previous studies on the effects of neuroticism on job stress did not analyze the effects of neuroticism on job stressors or the indirect effects on the stress response. A new finding of the present study is that the association between these factors have become clear, indicating that high neuroticism is associated with the sensitivity of perceived job stressors and at the same time is associated with the high psychological and physical stress response.

In this study, the quality of parenting was not associated with perceived job stressors directly, but was associated with them through neuroticism. In our previous study of general adults, the quality of parenting did not affect the negative appraisal of adult life events directly, which included job stress, but indirectly affected stressors only through its effect on neuroticism.¹⁸ On the other hand, the quality of parenting had a direct effect on PPSR. This suggests a possible pathway for the effects of the quality of parenting on PPSR through factors other than neuroticism and perceived job stressors (family, other personal factors, and social support). Interestingly, job stress of the parents has been reported to adversely affect parents’ attitudes toward their children, which adversely affect child adaptation indirectly.³⁶ In other words, job stress may transmit psychological adverse effects from parents to the next generation through parenting and neuroticism.

The results of this study suggest that improving the quality of parenting may lead to a reduction in perceived job stressors and PPSR of the next generation, by reducing neuroticism in children, and consequently reducing medical expenses and improving presenteeism and absenteeism. On the other hand, intervention of neuroticism is also expected to improve job stress. It has been reported that intervention by improving workplace bullying status reduces neuroticism, and mindfulness and cognitive behavioral therapy also reduces neuroticism.^37–39

Limitations

As this study is a cross-sectional study, we cannot conclude the causal association between the factors. Therefore, a prospective follow-up study is needed. Because the subjects in this study were recruited by convenience sampling, they may not reflect the general population (sampling bias). There is a possibility of sampling bias for evaluating only the subjects who gave consent for this study. The subjects who did not show consent for this study might be in worse or better condition of mental health compared to the participants. Subjective evaluation by the self-administered questionnaires used in this study may be different from objective evaluations. In addition, there is a possibility of recall bias as the subjects answered the questions by looking back on their childhood.

Conclusions

This study suggests that the quality of parenting is indirectly associated with the PPSR through its effects on neuroticism and perceived job stressors. Reducing neuroticism in children by improving the quality of parenting may reduce job stress in the next generation of workers, and intervention of neuroticism is expected to reduce job stress in the current generation of workers.