Background. Little is known of the preventive effects of physical activity in leisure time on neck and upper limb symptoms. Methods. A cohort of 1742 employees was selected from a prospective cohort study with a follow-up period of 3 years. Independent variables were sporting activities and physically active commuting. Outcome measures were neck/shoulder symptoms and elbow/wrist/hand symptoms as well as sickness absence due to these symptoms. To analyze the data, the generalized estimating equation (GEE) method was used, with adjustment for individual characteristics, such as age, gender, lifestyle, and the outcome at baseline. Results. Practicing sports for at least 10 months a year decreased the risk of neck/shoulder symptoms (OR: 0.82; CI: 0.67-0.99), sickness absence (OR: 0.48; CI: 0.28-0.84), and long-term sickness absence (OR: 0.37; CI: 0.17-0.84) due to neck or upper limb symptoms. A high mean intensity (≥3 h per week) of sporting activities had less effect than the continuation of these activities throughout the year. Conclusion. Sustained sporting activities have a favorable effect on neck/shoulder symptoms and on sickness absence due to neck or upper limb symptoms. An effect of physically active commuting could not be demonstrated, although there was a tendency towards a favorable effect on sickness absence. © 2004 Elsevier Inc. All rights reserved.