Background : Psychosocial factors at work (PFW) can be defined as all non-physicochemical occupational risks. Several epidemiological models have been proposed to measure PFW, but one of the most widely used is Karasek’s model.

Aims : To determine whether psychosocial factors, evaluated by Karasek’s questionnaire, had increased in a cohort of workers.

Methods : A random sample of workers in the Pays de la Loire region of France, who could be considered representative of the region’s population of salaried workers, filled in a self-administered questionnaire, including Karasek’s self-administered questionnaire, in 2002–05 and 2007–09. Karasek’s questionnaire can be used to study three psychosocial dimensions (psychological demand, decision latitude and social support in the workplace) in workers in order to define two high-risk situations for their health: ‘Job Strain’ and ‘Iso Strain’. Changes in job strain and iso strain among workers were studied according to the workers’ sociodemographic characteristics and their working conditions.

Results : In this sample of 2049 workers, the proportion with iso strain increased between the two periods from 12 to 16%, P < 0.001, mainly among manual workers. Deterioration of Karasek indicators was mainly explained by an increase of the ‘low social support’ dimension (38 versus 49%, P < 0.001). Working conditions such as temporary employment of colleagues and high perceived physical exertion were associated with higher PFW.

Conclusions : This study, based on a quantitative and collective model, showed deterioration of working team environments and increased risk for individual mental health in this cohort of French workers in recent years.