Risk of depressive symptoms among mining company workers: a psychosocial and gender approach
Elisa Ansoleaga* Programa de Estudios Psicosociales del Trabajo, Facultad de Psicología, Universidad Diego Portales, Chile
Juan Pablo Toro Programa de Estudios Psicosociales del Trabajo, Facultad de Psicología, Universidad Diego Portales, Chile
Ximena Díaz Centro de Estudios de la Mujer, Chile
ABSTRACT
In recent decades, new technologies, production processes and change driven by the global economy have profoundly transformed work in the mining industry. In addition to physical and environmental risks, workers face new psychosocial risks. Recognized psychosocial risk factors include psychological demands, job control, social support, effort-reward imbalance, organizational justice, and work-family conciliation. These factors are linked to emerging physical and mental pathologies, with occupational stress as a cross-cutting variable.
This study examines links between psychosocial risk factors and depression risk in a stratified by section, random sample of 303 mining company workers, 95 percent male. A questionnaire using scales probing for psychological demands, job control, social support, effort-reward imbalance, distress, drug consumption and depressive symptoms was administered. Results showed:
1)Higher depressive symptomrisks among workers with: low social support relative to those with high social support; high effort-reward imbalance relative to those with low effort-reward imbalance; high psychological demands relative to those facing low psychological demands; high or very high stress levels relative to those facing low stresslevels, and between those using more than one psychotropic drug relative to those who do not.
2)High prevalence of depressive symptoms, distress and psychotropic drug use in the sample, relative to the general male population.
Results are inconsistent with otherwise clean bills of general health, suggesting that male gender socialization hinders symptom and illness recognition. In addition, positive perceptions of work-family conciliation issues suggest over identification with male provider roles and scant connection with domestic work. The conclusion is that a gender approach can help shed light on links between health and working conditions.
______
* Corresponding author. Psychosocial Labour Studies Program, Faculty of Psychology, Diego Portales University, Chile, Vergara 275, Tel. (562) 676-2556,
INTRODUCTION
The organization of work has changed profoundly in recent decades (Neffa, 2003; Díaz et al., 2007; Ansoleaga, 2007), obliging businesses to adapt themselves to these changes in order to keep pace with increasingly higher standards of productivity and quality. Among the major changes in work and employment are a pronounced increase in the use of information and communication technologies, intensification of work in relation to the time available to carry out tasks or finish products and a rise in atypical employment, among others. These changes have a significant correlation with the health of workers, such that together with the recognition of physical-environmental risks, there is an awareness of a new type of work-related risk that is being termed psychosocial risk, that affects the health of individuals (Hurrel et al., 2001; Vézina, M., Derriennic, F. & Monfort, C., 2004). The study of the psychosocial aspects of work has acquired public relevance with the recognition that changes in the organization of work are placing emotional and cognitive demands on workers that translate into what is termed the psychic and mental load of work (Dejours, 1992; Parra, 2001, 2004; Vézina, 2007).
While there is very extensive literature on work-related psychological risks, there is consensus with regard to some of the basic causal factors, such as autonomy or control over work processes, psychological demands, the social support of superiors and peers, the effort invested and the recognition received in response and organizational justice. These factors are gathered under three recognized models, the demand-control model of Karasek and Theorell (1990), the effort-reward balance model of Siegrist (1996) and the organizational justice model of Moorman (1991). These models explain the mechanisms that produce stress or psychic strain associated with work. Evidence indicates that the sustained suffering of this strain over time acts as a precursor to physical and mental pathologies. Evidence indicates that there are causal relationships between psychic strain and cardiovascular (Fischer, 2003), musculoskeletal (Rebolledo et al., 1999) and gastrointestinal disorders (Drossman et al., 1999), among other physical effects. In relation to the association between psychological risk and mental pathology, relationships have been established between work demands and affective disorders (Bromet et al., 1988), inadequate work, poor human relationships, high psychological demands, low control and low social support in work with a high incidence of depressive symptoms (Niedhammer et al., 1998); and associations between psychosocial risks and poor mental health (Vézina, Derriennic & Monfort, 2004; Sánchez, 2006; Vézina, 2007).
Nevertheless, beyond the social-psychological risks associated with work conditions and labour relations, since the 1970s two theoretical bodies of thought have been developing that recognize the relevance of work-related psychosocial risk factors as determinants of mental pathologies among men and women workers. At the same time, gender studies have pointed to mental health risks associated with conflicts between work and family. Family life and professional life are more and more difficult to reconcile. Several changes in professional life and in family life are at the origin of more incompatibilities between these two spheres of life (Carlson & Frone, 2003; Carlson & Kacmar 2000). Work-family conflict has been shown to be related to negative work outcomes, such as job dissatisfaction, job burnout, and high turnover, as well as to outcomes related to psychological distress (Greenhaus, Parasuraman & Collins, 2001).
Many health scientists contend that health behaviours are among the most important factors influencing well-being. The most common theory in health education and promotion is the Health Belief Model (HBM), based on the concept that health behaviour is determined by personal beliefs or perceptions about health and disease. Many sociocultural factors are associated with health behaviour, but gender is among the most important (Courtenay, 2000; Macintyre, 1993; Cameron & Bernardes 1998). Some health-related perceptions, beliefs and behaviours that can be used in the demonstration of hegemonic masculinity include the denial of weakness or vulnerability, emotional and physical illness, the appearance of being strong and robust, and dismissive of any need for help. By dismissing their health care needs, men are constructing gender.
This is particularly relevant in the context of a sustained increase globally in the prevalence of mental health disorders (World Health Organization, 2009). Chile is not immune to this tendency. Between 2007 and 2009, medical leave because of “mental and behavioural disorders” increased by 63.1% (Boletín Estadístico Fondo Nacional de Salud, 2008-2009.) Pathologies associated with this increase are depression, anxiety and stress among the working age population.
However, there have been few studies in Chile about psychosocial risks and mental health associated with changes in employment and work conditions. In particular, large companies in the mining sector evaluate the level of perceived well-being of workers through organizational climate studies that do not allow for establishing relationships between perceived well-being and the workers’ health. The psychosocial perspective on which this study is based provides relevant information about specific dimensions of jobs and allows for identifying and managing risk factors and elements that protect workers’ health, with proven results on well-being, work satisfaction and workers’ health.
METHODOLOGY
Research questions and working hypothesis
Our research questions were designed to determine i) if there are differences in the risk of suffering depressive symptomatology according to the level of exposure to psychosocial risk at work; ii) if there are differences in the risk of suffering depressive symptomatology according to the perceived level of work/family conflict, and iii) if there are differences in self-perception of health and medical consultations among subjects that present a high level of depressive symptomatology. The working hypotheses are:
H1: Subjects with high psychological demand and low control have more risk of suffering depressive symptomatology than those with a low psychological demand and high control.
H2: Subjects with a high effort/reward imbalance have more risk of suffering depressive symptomatology than those with a low effort/reward imbalance.
H3: Subjects with low social support from their superiors have more risk of suffering depressive symptomatology than those with high social support from their superiors.
H4: Subjects with low social support from peers have more risk of suffering depressive symptomatology than those who have high social support form their peers.
H5: Subjects with high or very high levels of stress have more risk of suffering depressive symptomatology than those with low or moderate levels of stress.
H6: Subjects that use one or more psychotropic drug have a higher risk of suffering depressive symptomatology than those who do not use psychotropic drugs.
H7: Subjects with a negative perception of the work/family reconciliation report a more negative perception of their general health than those who have a more positive perception about the reconciliation.
H8. Subjects that present depressivesymptomatology, stress and the use of psychotropic drugs do not have a negative perception of their health and do not consult physicians more often than subjects who do not present these symptomatologies.
Design and sample
The design was a descriptive and associative crosscutting study of an exploratory character. It was conducted with employees of a company in the Chilean mining sector that has a workforce of approximately 4,800 permanent workers and another 7,000 sub-contracted workers. The sample consisted of a group of permanent employees on a shift work system. From a base of 4,853 subjects, a sample of n= 303 was obtained by random selection, stratified according to the size of the different areas of the company so as to provide representativeness. As a result, the sample was overwhelmingly male (95%), because of which no sex-stratified analysis was made. The ages of subjects ranged from 19 to 64 years of age and 50% of the sample were between 49 and 64 years of age.
Fieldwork
Data collection instrument
An adapted form of the EQCOTESST questionnaire (Institut de Recherche Robert-Sauvé en Santé et en Sécurité du Travail, 2010) was applied. Scales for psychological demand, control over work, effort-reward balance, stress and depressive symptomatology were used. The questionnaire also included four questions regarding access to company family/work reconciliation policies. Finally, to evaluate the relationship between perceived health and the symptomatology of mental health, a question was included about perceived state of health and two questions about medical consultations applied to subjects who answered affirmatively to the question about mental health symptomatology.
A pilot version of the survey was applied to fifty subjects. Following this some questions that had presented problems of comprehension were rewritten.
The instrument was applied between June and September 2008. Trained surveyors oversaw the application of the self-administered surveys that included signing an informed consent form and an expression of willingness to participate on the part of the subjects. The research project received the approval of the Ethics Committee of the sponsoring institution.
Statistical analysis
Points from low to high were assigned to a Likert-type scale with four response alternatives. Subsequently, new dichotomised variables were generated for each scale, in which the value of one indicated the presence of a risk factor and zero indicated the absence. For the scale of stress (K6), which had six items in a Likert scale, a range of four possibilities was established: low stress, moderate stress, high stress and very high stress. The scale for the use of psychotropic drugs had a score of 1 for persons who use at least one drug and 0 for persons who do not use any.
The questions on work/family reconciliation had two alternative responses (1: Yes, 2. No). The question on the perception on the state of health had four alternative responses (1: very good to 4: very bad).
Descriptive statistics and statistical measure of association were used. Data were processed with STATA 11.0 software.
RESULTS AND DISCUSSION
We found that 40% of the subjects presented a high psychological demand, 11% low control over their work process and 5% met the conditions to experience psychic strain. This represents low control and high psychological demand.
Some 33% of subjects experience an imbalance between the effort invested and the rewards received, 22% experience a low level of social support from their superiors, 8% experience a low level of support from their peers and 12% present high or very high levels of stress. With regard to the use of psychotropic drugs, 27% declared regular use of sedatives; 20% use hypnotics and 14% anti-depressants. Some 34% reported having used at least one of the three types of psychotropic drugs. Likewise, 23% responded affirmatively to the question about depressive symptomatology.
The subjects who received low social support from their superiors had a 79% higher risk of presenting depressive symptomatology than those who received high social support from their superiors. Consequently, we accept the hypothesis that there are differences in the risk of suffering depressive symptomatology according to exposure to a low level of social support from superiors. There were no significant differences according to exposure to a low level of support from peers.
Subjects who report an effort-reward imbalance have a 78% higher risk of suffering depressive symptomatology than those reported a good effort-reward balance. Consequently, we accept the hypothesis that there are differences in the risk of suffering depression according to exposure to an effort-reward imbalance.
There were no statistically significant differences between subjects who have high psychic strain and those with low psychic strain in terms of the risk of suffering depressive symptomatology (p = 0.747). Consequently, we accept the null hypothesis that there are no differences in the risk of suffering depression according to the level of exposure to psychic strain. On the other hand, subject with high or very high levels of stress are four and a half time more likely to suffer depressive symptomatology than those who with moderate or low stress. Therefore, we accept the corresponding hypothesis.
The subjects who use one or more psychotropic drug have four times as much risk of suffering from depressive symptomatology as those who do not use psychotropic drugs. Consequently, the hypothesis is accepted that there are differences in the risk of suffering depression according to use of psychotropic drugs.
In relation to the work/family reconciliation, the subjects reported a very positive perception about business policies: 23% can choose working hours or exchange schedules with co-workers, 62% have the right to paid leaves to be absent from work for family matters; almost 80% can deal with personal matters during working hours; 77% can ask for unpaid leave from work. There are no significant differences in relation to the risk of suffering depressive symptomatology among subjects who have a positive perception of the employer’s policies of reconciliation and those who have a negative perception. Consequently, the corresponding hypothesis is not accepted.
The results in relation to the perception of general health show that 69% of the subjects reported that their general state of health was good or very good. However, 51% of subjects that responded positively to the two questions about depression (loss of interest in most things and a continuous feeling of sadness), and 42% of the subjects that reported high or very high levels of stress on the K6 scale, indicated that their health was good or very good. Consequently, of the 125 subjects who responded affirmatively to mental health symptomatology, only 42%, or 55 subjects, had consulted with specialists. The hypothesis is thus accepted that subjects who present symptomatology of depression, stress and the use of psychotropic drugs do not report a poorer state of health than those who do present these symptomatologies, nor do they consult physicians more on account of this.
Several aspects of the results deserve discussion. Firstly, the strong correlation between depressive symptomatology and/or stress and the use of psychotropic drugs is very understandable given that it is highly probable that individuals who feel depressed or stressed use psychotropic drugs. It is also important to remember that the situations of stress are often the precursor of the emergence of mental pathologies, because of which it is completely reasonable that these variables are closely associated. The point of interest lies precisely in the prevention of depressive symptomatology among those who present high or very high levels of stress.
Another point of interest is the high prevalence in the use of psychotropic drugs among our sample. For example the figures for the use of anxiolytics among the general male population of Greater Santiago (between 19 and 64 years of age) (Fritsch et al., 2005) is 1.95%, while in our sample it reached 27%. Some 20% of our sample uses sedatives, while the figure for the general population is 0.19%. Likewise, 14% of our sample uses antidepressant drugs, while the figure for the general population is 1.69% (Rojas et al., 2005).
Finally, the discordance between the perception of general health and the data on depressive symptomatology and the prevalence of stress is noteworthy. As can be appreciated from the results, there is marked discrepancy between the positive report on the state of health on the one hand, and on the other hand, the high or very high levels of stress, the positive responses to indicators of depression and the low number of consultations with specialists.
This can be interpreted as a manifestation of the invisibilization of mental health in our national context and particularly in the context of a hyper-masculinized work culture, as is the case of mining in Chile. As we have discussed before, gender is one of the most influential on beliefs and perceptions about health. In order to follow any positive health behaviours, men must reject multiple constructions about masculinity. On the other hand, as one researcher found, in an effort to preserve their masculinity, men with chronic illnesses often work diligently to hide their disabilities (Courtenay, 2000). Men are more inhibited than women about reporting certain conditions that are less compatible with culturally acceptable male roles (Macintyre, 1993) and are less likely than women to share their health concerns with others, including health care professionals (Koopmans & Lamers, 2007). Men also tend to use primary health services less than women and are more likely to delay seeking help when they are ill. As well, they are more likely to adopt health damaging or risky behaviours (Cameron & Bernardes, 1998).