A cross-sectional study using data from the 2007 household survey in England has founda strong association between work stress and common mental health disorders, which is independent of the effects of non-work stressors.¹

The 3,383 participants (1,804 men) were of working age, in employment and not on long-term sickness absence. Common mental health disorders were diagnosed using the clinical interview schedule (CIS-R). This is a structured interview which enquires about the presence and severity of 14 common psychiatric symptoms during the week before the interview.

Computer algorithms were used to define ICD-10 diagnoses which were then grouped into the six categories: mixed anxiety and depression, generalised anxiety disorder (GAD), depressive episode, phobia, obsessive compulsive disorder and panic disorder.

Two models of work stress were used. According to Karasek, job strain results from high demands coupled to lack of perceived control, and is made worse if there is a lack of support from colleagues.² According to Siegrist, stress results from an imbalance between effort and reward, and is more likely among those who are excessively committed to their work.³

Questionnaires were used to assess both work and non-work stressors. For each model of work stress, participants were assigned to one of four categories: low demands, high control (no strain); high demands, high control (active job); low demands, low control (passive job); high demands, low control (job strain).

After adjustment for sociodemographic factors and social support at work, job strain was associated with a trebling of the odds of common mental health disorders, when compared with the no strain group, odds ratio (OR) 2.99 (95% CI: 2.12-4.24). A lack of social support at work was also significantly associated with common mental health disorders, OR 1.71 (95% CI: 1.32-2.23).

After adjustment for sociodemographic factors and overcommitment, effort reward imbalance (low rewards, high effort) was also associated with a trebling of the odds of common mental health disorders, when compared with the high rewards, low effort group, OR 3.03 (95% CI: 2.17-4.24). Overcommitment was found to have the strongest association with common mental health disorders, OR 3.52 (95% CI: 2.57-4.81).

Adjustment for non-work stressors had very little effect on the results, indicating that work stressors have an independent effect on common mental health disorders.

It would be of interest to investigate the link between work stress and specific psychiatric conditions, but this would have required a much larger sample size. More than half of those with a common mental health disorder (55%) had mixed anxiety and depression (MADD) i.e. subthreshold depressive and anxiety symptoms.⁴ The status of ICD-10 MADD is questionable: it is poorly defined and usually self-limiting, but is associated with significantly impaired quality of life and high levels of work absence.⁵

A cross-sectional study cannot demonstrate that work stress causes mental ill health. Alternative explanations for the association could be that individuals who are at increased risk of developing common mental health disorders are more likely to choose stressful jobs, or that depressed patients are more likely to perceive their work in a negative way. The Dunedin longitudinal study attempted to control for both of these possibilities and found that excessive job demands were associated with a doubling of the risk of major depression or GAD.⁶

Doctors have the highest incidence of work-related mental ill health of any profession.7 This may be the consequence of high demands (e.g. increasing complexity of consultations) and diminishing autonomy (guidelines and performance measures). Doctors who have an excessive need for approval and esteem from work (overcommitment) are particularly vulnerable. Conversely, doctors who work within supportive partnerships will have a reduced risk. In my experience, 360° appraisal can facilitate positive feedback and help colleagues and staff to feel both valued and supported.