Early-life physical activity and risk of juvenile idiopathic arthritis: a longitudinal ABIS cohort study using a symptom-free baseline design

Abstract

The role of physical activity (PA) in the development of juvenile idiopathic arthritis (JIA) remains unclear, and the evidence addressing this question is limited. Therefore, this study aimed to investigate the longitudinal association between early-life PA and the risk of JIA, using a symptom-free baseline design to minimize bias from early disease manifestations. A total of 16,415 participants were included from the All Babies in Southeast Sweden cohort. PA was assessed at age 5, 8, and 10–12 years using a composite Total Activity Score by categorizing participants into less active and highly active groups based on z-score threshold. Incident JIA cases (n = 88) were identified through Swedish National Patient Registers. At each age, individuals with a diagnosis of JIA prior to or at the time point were excluded. Participants reporting joint symptoms (pain, swelling, limited mobility, or limping) were excluded to create a symptom-free baseline population. Remaining participants were followed for incident JIA. Cox proportional hazard models were used to estimate hazard ratios (HRs) with 95% confidence interval. All models were adjusted for sex, parental education, and family history of rheumatic disease. After exclusions, 6832 participants were identified at age 5, 3885 at age 8, and 4445 at age 10–12. A total of 62, 56, and 40 incident JIA cases during follow-up were identified in the respective age-specific cohorts. Higher PA was consistently associated with a reduced risk of JIA across all age groups, with HRs ranging from 0.72 to 0.88 in unadjusted models, and 0.76 to 0.91 in adjusted models. However, none of the associations reached statistical significance. Our findings suggest that PA in childhood does not increase the risk of JIA although any potential protective association is likely modest. The use of a symptom-free cohort strengthens causal interpretation by minimizing reverse causation. • To investigate whether higher PA in early childhood is associated with the risk of developing JIA, while accounting for potential bias from early joint symptoms. • In a large prospective birth cohort, higher PA was associated with a consistent inverse trend in JIA risk across multiple age groups; however, these associations were not statistically significant when analyses were restricted to symptom-free individuals. • Our findings suggest that PA is unlikely to have a strong independent effect on JIA risk but may play a modest role. The study highlights the importance of accounting for early disease manifestations in epidemiological research. Further large-scale studies might needed to clarify the role of PA in the development of autoimmune diseases in children.