The Bacillus Calmette–Guérin (BCG) vaccine is one of the most widely used vaccines globally. In 2005, England changed from universal vaccination of school-age children to targeted vaccination of high-risk children at birth. The impact of this policy change has not been evaluated. We combined notification data from the Enhanced Tuberculosis Surveillance system, with demographic data from the Labour Force Survey to construct retrospective cohorts of individuals in England eligible for both the universal, and targeted vaccination programs between Jan 1, 2000, and Dec 31, 2010. For each cohort, we estimated the incidence over a 5 year period and used Poisson regression models in order to estimate the impact of the change in policy on TB burden in England. We found evidence that ending universal vaccination was associated with a 12% (95%CI 3-21%) increase in incidence rates in the UK born school-age population eligible for the programme during the study period study for the univariable model, and of 17% (95%CI -7-47%) in the fully adjusted model. We found more variable evidence for an association between the high-risk neonatal program and reduced incidence rates in UK born neonates eligible for the programme during the study period; with a 19% (95%CI -1-35%) reduction in incidence rates in UK born eligible neonates in the fully adjusted model, but little evidence of any association in the univariable model. The introduction of the targeted BCG vaccination programme and withdrawal of universal vaccination was associated with increased incidence rates in those at school-age and reduced incidence rates in neonates. Understanding the trade-offs inherent to vaccination targeting could inform future decision-making. The authors are funded by the National Institute for Health Research Health Protection Research Unit (NIHR HPRU) in Evaluation of Interventions at the University of Bristol in partnership with Public Health England (PHE).