Background: The aetiology of systemic lupus erythematosus (SLE) remains unclear. Clinical observations and a small number of studies performed so far suggest an association between psychological stress and self-reported symptoms of SLE patients. This longitudinal study was designed to investigate whether daily psychological stress is associated with flares in SLE patients, measured by clinical and laboratory parameters. Methods: Female SLE patients (n = 41) were followed over a period of six months. Daily stress was monitored by a hand-held PC diary programmed with 44 items based on standardized measures and clinical experience. Once every four weeks patients visited the outpatient clinic for medical evaluation. Disease activity was evaluated using the European Consensus Lupus Activity Measurement (ECLAM), laboratory parameters, and intake of steroids. Results: Classification and regression tree (CART) patient-wise analyses revealed that SLE patients with vs. without flares using complement and ECLAM as activity measures show greater negative self-ratings in mood, and social duties (p < 0.01). In addition, mixed model analysis of variance showed that daily hassles with social relationships were significantly associated with flares in SLE measured by an increase in steroid medication >5mg/d (p < 0.01). Conclusions: These results suggest that psychological stress is associated with flares in SLE. Particularly daily stress with social relationships and social duties may be factors to be related to the course of disease activity in SLE.