Anxiety disorders are found to be the most prevalent psychological problems among children and adolescents. Cognitive behaviour therapy (CBT) was found to be effective at reducing anxiety. The purpose of this study was to assess the effectiveness of a universal school-based intervention for reducing anxiety among Grade 9 schoolchildren. A randomized controlled cluster trial was conducted by randomly assigning 36 schools in the Colombo district in Sri Lanka into study and control arms, each comprising 18 schools with 360 students. The levels of anxiety, self-esteem and depression status were assessed using the validated Screen for Child Anxiety Related Disorders (SCARED) tool and the Rosenberg self-esteem scale, respectively, and the Depression, Anxiety and Stress Scale 21 (DASS-21) at baseline, after intervention, and after 3 months. A CBT-based universal intervention package was administered weekly by a trained teacher for eight weeks with a one-month self-practice period to a randomly selected Grade 9 class in each school in the study arm. The control arm received routine care. Anxiety and self-esteem scores and depression status were compared between the two arms after the intervention and at 3 months of follow-up using the generalized estimation equation (GEE), controlling for confounding and clustering. The nonresponse and loss to follow-up rates were < 1%. When comparing the study arm with the control arm using GEE, anxiety levels were significantly lower [β = (-0.096), 95% CI = (-0.005) - (-0.186), p = 0.038] at follow-up but not postintervention [β = (-0.024), 95% CI = 0.006 - (-0.055), p = 0.115]. There were no significant differences in depression status after intervention (OR = 0.257, 95% CI =0.052-1.286; p = 0.098) or follow-up (OR = 0.422, 95% CI 0.177-1.008; p = 0.052), and self-esteem significantly increased after intervention (β = 0.811, 95% CI = 0.314-1.309; p = 0.001) but not at follow-up [β=0.435, 95% CI=(-0.276)-1.145, p=0.231]. This study revealed that the universal package based on CBT is effective at reducing anxiety and improving self-esteem among adolescents. The trial registration number and date were SLCTR/2018/018 and 19th of June 2018 respectively.