Background: The aim of this study was to indicate psychosocial stressors at work that significantly affect sickness absence among workers. Materials and Methods: Study subjects included a group of 233 randomly selected women employed as post-office clerks. Sickness absence data covered the period of 2004-2006. The psychosocial factors were assessed by means of the Subjective Work Characteristics Questionnaire. The hazard ratio (HR) of sickness absence was analysed using the Cox regression model, separately for short- (1-9 days) medium- (10-29 days) and long-term (30 days and above) sickness absence. Results: The shortterm sickness absence risk was significantly related with the post-office size - in the offices employing 8-12 workers, the risk was by 50% lower compared to those employing a smaller number of workers (HR = 0.49; 95% CI: 0.27-0.90) and unpleasant working conditions (dirt), which contributed to the increased risk (HR = 1.30; 95% CI: 1.12-1.50). In the case of a 10-29-day absence, the risk was slightly elevated by the demand of long-term vigilance, financial responsibility, and strictly determined breaks at work. In the model of long-term sickness absence, a significantly higher risk was noted when the number of employees was 16-25 compared to a smaller number of employees (HR = 2.92; 95% CI: 1.09-7.82), non-occupational, self-assessed workload was high (HR = 2.97; 95% CI: 1.34-6.62) or moderate (HR = 2.22; 95% CI: 1.11-4.44) compared to self-assessed low workload, and the work space was limited (HR = 1.21; 95% CI: 1.00-1.47). Conclusions: Our analysis showed a significant effect of stressogenic work conditions on the patterns of sickness absence. Our findings may help in developing programs intended to reduce sickness absence through limiting the prevalence of unfavourable conditions at workplaces. Med Pr 2009;60(4):259-271