Background Occupational and leisure-time physical activity are considered to provide similar health benefits. The authors tested this hypothesis.

Methods A representative sample of Danish employees (n=7144, 52% females) reported levels of occupational and leisure-time physical activity in 2005. Long-term sickness absence (LTSA) spells of ≥3 consecutive weeks were retrieved from a social-transfer payment register from 2005 to 2007.

Results 341 men and 620 females experienced a spell of LTSA during the period. Cox analyses adjusted for age, gender, smoking, alcohol, body mass index, chronic disease, social support from immediate superior, emotional demands, social class and occupational or leisure-time physical activity showed a decreased risk for LTSA among workers with moderate (HR 0.85, CI 0.72 to 1.01) and high (HR 0.77, CI 0.62 to 0.95) leisure-time physical activity in reference to those with low leisure-time physical activity. In contrast, an increased risk for LTSA was shown among workers with moderate (HR 1.59, CI 1.35 to 1.88) and high (HR 1.84, CI 1.55 to 2.18) occupational physical activity referencing those with low occupational physical activity.

Conclusion The hypothesis was rejected. In a dose–response manner, occupational physical activity increased the risk for LTSA, while leisure-time physical activity decreased the risk for LTSA. The findings indicate opposing effects of occupational and leisure-time physical activity on global health.