There is a growing evidence base that supports the association between workplace stress and the development of CMDs.

Stress in the workplace may have a pervasive effect on employees, leading to exhaustion, anxiety and depression, and even substance abuse. Repeated changes in the workplace can precipitate additional stress. It is well recognised that stress contributes to high levels of absenteeism in the workplace. Stressors have been defined as a set of circumstances which have an adverse impact on a person's equilibrium. This equilibrium is also influenced by the individual's coping strategies and resources, which are inevitably dependent on the person's environment. Stressors may include various factors such as job insecurity, hazardous working conditions, high workload, the threat of violence, unrealistic deadlines, lack of managerial support and retribution from complaints procedures. Other social factors also have an impact on work productivity, including interpersonal relationship difficulties, loss and physical illness. It should also be acknowledged that personality profiles, lower levels of personal resources and lower resilience may also be associated with an increased vulnerability of developing CMDs.

Workplace culture is a mediating factor in either reducing or increasing stress. Morale, autonomy and team dynamics can have an effect on workplace stress and subsequently productivity. Bullying, discrimination and abuse of employees are extreme examples of poor workplace culture. On the other hand, investment in creating a positive workplace culture can be viewed as an investment in social capital, which is a resource that all individuals can access.

The risk of CMDs is higher in workplaces characterised by a high pace of work and low skill discretion. In general, unskilled workers are reported to have a higher risk of CMDs compared with white collar workers. For female employees in particular, lack of job autonomy and decision-making procedures are risk factors for CMDs.

Using the demand-control model, there is evidence that jobs characterised by high demands with respect to workload, time pressure and role conflict increase the risk of psychiatric morbidity. Furthermore, workers with low autonomy and authority are most vulnerable, particularly those who have limited external social support. In a survey of more than 1.000 Victorian workers LaMontagne et al (2008) demonstrated a clear correlation between job strain and depression. The population-attributable risk was 13.2% for males and 17.2% for females.

An imbalance in the effort-reward paradigm has also been associated with an increased risk of psychiatric morbidity. Tsutsumi and Kawakami (2004) recommend redressing the effort-reward imbalance through encouraging employee control over work scheduling tasks and responsibilities, as well as improving rewards, developing additional reward schemes, supervisor training in the maintenance of a positive relationship with employees, and providing incentives to employees for career development.

Several other studies have documented an association between workplace stress, defined in various ways, and depression. The type of employment contract may significantly affect psychiatric morbidity. In particular the British Household Panel Survey found an association between precarious employment and psychiatric morbidity, with a significant longitudinal association demonstrated for men. It could be postulated that work security and lack of reward opportunities in relation to the degree of effort can be a potential source of stress. Kawakami et al (1990) found in a study of male industrial workers in Japan that jobs associated with high levels of stress had a more than 11-fold relative risk of depression. Virtanen et al (2007) used antidepressant prescription as a proxy measure for depression in a study of Finnish workers and found a positive correlation between job-strain and depression.

Low social support at work has been shown to be associated with an increased risk of depression. In the NEMESIS study, a high degree of social support was negatively associated with depression, with a relative risk of 0.8. Other studies have confirmed that low social support, including coworker and supervisor support, are associated with an increased risk of depression.

Unique work exposures are of course associated with a higher risk of developing CMDs. For example, Fullerton et al (2004) found that rescue workers exposed to physical danger had a relative risk of developing depression of 3.5 compared with the rest of the population. In an interesting study by Berg et al (2006) of police officers in Norway, particular factors associated with CMDs were identified that could be considered as common to other occupations. These included job pressure and lack of support. Other factors were identified that were specific to the occupation of police work; frequent work injuries were not surprisingly associated with an increased rate of depressive symptoms. Although post-traumatic stress disorder was not specifically investigated, police reported more depersonalisation in comparison with the general population control group.

Zammuner and Galli (2005) noted the impact of emotional labour, the act of expressing emotions that are desirable for the organization, which can place a significant burden on the employee's emotional well-being. This occurs as a result of the stress associated with regulating emotions during interactions in the workplace that may be stressful. Emotional labour was associated with burnout. Whilst this is relevant to other workplace settings, the impact of emotional labour in impoverished workplace settings is likely to be magnified.

In a study be Wall et al (1997), cited by Munn-Giddings et al (2005), 11,637 employees of the UK National Health Service (NHS) were interviewed and a high level of psychological distress was found amongst this cohort. This has significant implications as health service staff are in the position of being professional carers, and their role may be compromised by their own mental well-being. The primary stressors faced by employees of health services include lack of resources and dysfunctional team dynamics, rather than the burden of caring for individual patients.

In another study of NHS employees in the UK, Loretto et al (2005) demonstrated that there are a wide range of personal, environmental as well as workplace factors which influence the well-being of employees. Conflict between work and non-work activities has a significant adverse impact. Support from management and a sense of autonomy were positively associated with wellbeing whilst high work demands and numerous changes at work had a negative impact. Loretto et al (2005) found that work pressure is associated with work-life imbalance which in turn has an adverse impact on psychological health, with an increased likelihood of employees suffering from a diagnosable CMD.

There have been relatively few studies that may be more relevant to the majority of workers in developing countries. Using the Hopkins Symptoms Checklist (HSCL-25) in a study of 374 female cleaning personnel in Norway, Gamperiene et al (2006) found that 17.5% of all personnel had evidence of a CMD. This figure is more than double the average prevalence of CMDs among working women in Norway of 8.4%. The cleaning profession was chosen by the authors as this occupation is known to be associated with several risk factors for stress in the workplace including low pay, lack of esteem and lack of control over working conditions. Poor satisfaction with leadership and poor satisfaction with co-workers were significantly associated with poor mental health. Interestingly, shift work and job strain were not found to be associated with mental health problems in this study. Cleaning staff who were immigrants were three times more likely to have CMDs compared with staff born in Norway. Also, employees in the 50–59 age bracket had a higher prevalence of mental health problems compared with younger employees as well as older employees approaching retirement.

It is interesting to note that despite evidence for this association between workplace stress and CMDs, this is usually not sufficient for affected employees to receive compensation. This is due to several factors, including the view that the association may not be proven independent of other stressors. Compensation courts often view depression as a condition that cannot be proven because it does not have any objective signs. Furthermore, compensation courts are wary of potential malingerers. LaMontagne et al (2008) also noted that depression associated with job strain is most probably under-recognised, as there are fewer numbers of individuals seeking compensation as a result of job strain.