OBJECTIVE: To examine the factors associated with back pain within a working population. METHODS: A cross-sectional survey of employees using a self-administered questionnaire and physical fitness tests. This study assessed 10,321 participants (6,251 male and 4,070 female) of two nationwide companies throughout Switzerland between 1996 and 1998. The participation rate was 41%. RESULTS: Of the participants, 4,945 (48%) suffered mild back pain and 696 (7%) suffered severe back pain. Reported "stress" was associated with back pain of any intensity. Abdominal muscle strength was inversely associated with severe back pain, while physical activity was non-linearly associated with severe back pain. Smoking was directly associated with any intensity of back pain. The variables gender, obesity, strength of the abdominal musculature and frequency of physical activity were insignificant for back pain of any intensity in multivariate analyses. CONCLUSION: Our results confirm the association of back pain with physical and behavioural factors. The non-linear relationship between physical activity and back pain may need further examination. Performing any kind of sport three to four times a week appears optimal.