Evaluating the effectiveness of the human papillomavirus vaccination programme in England, using a regression discontinuity design

Abstract

Background In England, the national human papillomavirus (HPV) immunization programme was introduced in 2008 to prevent cervical cancer. Girls aged 12–13 years were offered routine vaccination and those aged 14–18 years in 2008 were offered ‘catch-up’ vaccination. We evaluate the effect of the HPV catch-up vaccination programmes on cervical cancer and cervical dysplasia diagnoses, and provide an estimate of the vaccine effectiveness. Methods Using the 2011 Census, Hospital Episode Statistics, and mortality data for the population of England, we exploit the cut-off in eligibility and apply a regression discontinuity design to assess the impact of HPV vaccination on cervical disease. Results Vaccination reduced the incidence of cervical dysplasia and cancer diagnoses by 31% and 75%, respectively, at ages 23–30 years in girls offered catch-up vaccination at ages 17–18 years compared with those who were just above the eligibility age for the catch-up vaccination, with a clear discontinuity. Reductions continued amongst girls offered routine vaccination. Conclusion These estimates, obtained by using a quasi-experimental approach, are similar to vaccine effectiveness estimates based on more traditional approaches. This approach provides further evidence of the HPV vaccination programme reducing adverse cervical outcomes in young women and could be used for future studies to evaluate major changes in HPV vaccination policy and for studies of longer-term outcomes including other cancers and deaths.