OBJECTIVES: Children with JIA can experience delayed and restricted growth. The objectives of this study were to investigate the influence of etanercept (ETN) on vertical growth and factors associated with improved growth in patients with JIA over the initial 2 years of treatment. METHODS: This analysis was restricted to ETN-treated patients in the British Society for Paediatric and Adolescent Rheumatology Etanercept Cohort Study with complete height data recorded at baseline, 1 and 2 years. Height z-scores were calculated using the World Health Organization growth standards for age and gender. Change in height z-score was evaluated over time. Multivariable linear regression was used to identify factors associated with change in height z-score from baseline. RESULTS: A total of 191 ETN-treated patients were included: 65% female, median baseline age 11.0 years [interquartile range (IQR) 7.3-12.9], median disease duration 3.5 years (IQR 1.7-7.1). At baseline mean height z-score was -0.74 (s.d. 1.4). After 2 years mean height z-score increased to -0.45 (1.4) (change +0.29; P <0.001). In multivariable analysis, factors associated with an improvement in height z-score included lower baseline height z-score [-0.110 per unit (95% CI -0.161, -0.059), P <0.001] and no oral corticosteroid use at baseline [-0.192 (95% CI -0.343, -0.040), P = 0.013]. CONCLUSION: ETN therapy was associated with an improvement in height z-score over the first 2 years of therapy in this real-world cohort of children with severe JIA. The lack of a strong association of improvements in height with improvements in disease activity warrants further exploration.