Psychosocial work environment and hospital admissions due to mental disorders: A 15-year prospective study of industrial employees

Abstract

Background

Low levels of job control and social support may increase the risk of mental disorders, particularly depression, but the evidence is mostly based on self-reports. We examined whether components of job control and work-related social support predict medically-certified mental disorders.

Methods

13 868 forest company employees with no previous hospital admissions for mental disorders responded to questionnaires on decision authority, skill discretion, co-worker and supervisor support. They were followed-up for hospital admissions due to mental disorders (ICD-9 codes 290 to 319), using national hospital discharge records (577 hospitalized, mean follow-up 15.1 years).

Results

In analyses adjusted for confounders, high skill discretion was associated with a reduced risk of hospital admission for mental disorders (HR 0.74, 95% CI 0.58–0.95). High decision authority was associated with an elevated risk (HR 1.48, 95% CI 1.17–1.87). Diagnosis-specific analyses showed high skill discretion to associate with a reduced risk of both depressive and non-depressive non-alcohol-related mental disorders. High decision authority was a risk factor for alcohol-related and depressive disorders. Good co-worker support was associated with a reduced risk of non-depressive non-alcohol-related mental disorders. Supervisor support was not associated with any mental disorders.

Limitations

We used a single time point estimate in an industrial sample comprising largely of men.

Conclusions

Contrary to previous research on job control, high decision authority increased the risk of depressive and alcohol-related disorders, which suggest a need to reconsider the strategies for prevention and clinical practise in regard to psychosocial work environment and mental health.

Introduction

Mental disorders, particularly depression, have increased dramatically in recent years as the cause of work disability and early retirement (Järvisalo et al., 2005). The financial cost of depression alone was estimated at 118 billion euro in Europe in 2004, accumulating directly from the usage of the health care system and indirectly due to sickness absences and early retirement (Sobocki et al., 2006). Poor psychosocial work environment causing stress is believed to be a risk factor for mental disorders but definite evidence is still missing.

Various adverse psychosocial work characteristics have been shown to prospectively predict mental ill health. In recent meta-analyses, job strain, low job control, poor social support, high psychological demands, effort–reward imbalance and high job insecurity have all been associated with approximately 20 to 80 percent excess risk of depression and mental ill health in general (Netterstrøm et al., 2008, Bonde, 2008, Stansfeld and Candy, 2006). The meta-analyses, however, listed various limitations: data on the associations between psychosocial factors and mental ill health are available from only 2 to 5 independent samples per work characteristic, and analyses show signs of publication bias. The follow-up period in most prospective studies on mental ill health has been limited to 1 to 2 years. Moreover protection against common method bias has been weak, as both exposure and outcome have typically been measured using questionnaires. In recent studies where either the work environment evaluations or the mental health outcome have been derived by means other than self-reports, psychosocial risk factors have shown little effects on mental health (Bonde et al., 2009, Waldenström et al., 2008, Wieclaw et al., 2008).

Also studies measuring psychosocial work characteristics do not sufficiently differentiate between dimensions within broader constructs. For example, dimensions of job control, such as decision authority and skill discretion, could contribute differently to health outcomes (Bosma et al., 1998), and social support can have unique qualities depending on whether it is in relation to colleagues or the supervisor (Väänänen et al., 2003, Sinokki et al., 2008). Moreover, many of the longitudinal studies so far have used public sector cohorts which may limit their generalizability to other settings. Also, in studies where the mental health outcome is based on general questionnaire measures of mental ill health (e.g. on the General Health Questionnaire, Stansfeld et al., 1999) a further interpretation regarding the clinical relevance is needed.

This study seeks to examine the associations between adverse psychosocial work characteristics and mental ill health among industrial employees in a robust prospective design that includes a clinical measurement of mental health, i.e. hospital admissions with specific diagnoses, with a mean follow-up of 15.1 years.